Mass Solitary and Mass Incarceration: Explaining the Dramatic Rise in Prolonged Solitary in America’S Prisons

Home , Robert Hillary King, Solitary (Lost)

MASS SOLITARY AND MASS INCARCERATION: EXPLAINING THE DRAMATIC RISE IN PROLONGED SOLITARY IN AMERICA’S PRISONS

Jules Lobel

ABSTRACT—In the last two decades of the twentieth century, prisons throughout the United States witnessed a dramatic rise in the use of solitary confinement, and the practice continues to be widespread. From the latter part of the nineteenth century until the 1970s and ’80s, prolonged solitary confinement in the United States had fallen into disuse, as numerous observers and the United States Supreme Court recognized that the practice caused profound mental harm to prisoners. The reasons for this dramatic rise in the nationwide use of solitary confinement and the development of new supermax prisons have not been explored in depth. In particular, there has been little critical discussion of the rise of mass prolonged solitary as a product of the mass incarceration of the last several decades of the twentieth century. This Essay locates the rise of mass solitary in the 1980s in the context of mass incarceration. It explains the dramatic expansion of the use of solitary confinement and the construction of new super-maximum (supermax) prisons as an attempt by prison officials and politicians to maintain control of prisons in the face of increasingly radicalized, rebellious prisoners—often, but not exclusively, African-American—who had organized protests and disobedient conduct in American prisons from the 1960s to the 1980s. The rise of solitary was connected to the use of mass incarceration as a form of social control. As society became more violent, so too did many prisons, but to view that violence as the underlying cause of the growth of supermax and other segregated confinement obscures the deeper, underlying causes of the rise of mass solitary. Those causes are linked to the rise of mass incarceration itself. Uncovering the history and causes of the dramatic rise in supermax prisons and the use of prolonged solitary confinement in the 1980s and ’90s is critical to understanding not only how we got to where we are, but how we can end this cruel and inhumane practice. The first Part of this Essay recounts the origins of the supermax prison at Marion Federal Penitentiary in the late 1970s and early 1980s and

159 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W demonstrates that the rise of mass solitary was more an official reaction to the need to control politically active and disruptive prisoners than to the violence narrative. The second Part explores prison officials’ need to reassert control over their prisoners and draws the parallels between the rise of both mass incarceration and mass solitary as a racialized mechanism of social control. The third Part introduces the preventive paradigm as a model to control prisoners and demonstrates that the concept of preventing future misconduct fueled both mass incarceration and the modern supermax, resulting in minimizing due process restraints and erroneously isolating thousands of people. Finally, the last Part analyzes the current reform movement and the alternatives that have been proffered and utilized to replace solitary, supermax confinement. The Essay concludes that prolonged solitary confinement can be abolished, and that prison officials have alternatives that can safely manage even very dangerous prisoners.

AUTHOR—Bessie McKee Walthour Professor of Law, University of Pittsburgh Law School. I want to thank my research assistants Emma Carson and Ariel Fleischer for their invaluable research, and Professor Todd May and Staughton and Alice Lynd for their insightful comments and advice. I also thank Dean Amy Wildermuth of the University of Pittsburgh Law School for the financial support and encouragement for my research and the staff of the University of Pittsburgh Document Technology Center for their assistance in preparing this manuscript.

INTRODUCTION ...... 160 I. THE VIOLENCE NARRATIVE AND THE RISE OF THE SUPERMAX ...... 169 A. The Violence Narrative ...... 169 B. Marion and the Creation of the Modern Supermax ...... 172 II. SUPERMAX AND CONTROL OF DISRUPTIVE OR REBELLIOUS PRISONERS ...... 178 III. THE PREVENTIVE MODEL ...... 187 IV. ENDING PROLONGED SOLITARY CONFINEMENT ...... 195 A. Rejecting the Preventive Paradigm...... 196 B. Alternatives to Solitary Confinement for Violent Prisoners ...... 201 CONCLUSION ...... 208

INTRODUCTION In the last two decades of the twentieth century, prisons throughout the United States witnessed a dramatic rise in the use of solitary confinement, and the practice continues to be widespread. From the latter part of the

160 115:159 (2020) Mass Solitary and Mass Incarceration nineteenth century until the 1970s and ’80s, prolonged solitary confinement in the United States fell into disuse, as numerous observers recognized that the practice caused profound mental harm to prisoners.1 In 1890, the United States Supreme Court summarized the mental harm caused by solitary confinement, noting that “[a] considerable number of the prisoners fell . . . into a semi-fatuous condition, from which it was next to impossible to arouse them, and others became violently insane, others, still, committed suicide” and even “those who stood the ordeal better . . . in most cases did not recover sufficient mental activity to be of any subsequent service to the community.”2 The era of large-scale isolation practiced in the early nineteenth century thus came to an end in the beginning of the twentieth century.3 Isolation was still used in American prisons, but typically as short-term punishment and on a much smaller scale.4 Even the harshest prison in the federal system, the infamous and widely criticized Alcatraz—which made no pretense of rehabilitation, employed no teachers, social workers, or psychologists, and severely limited contact with the outside world—nonetheless provided congregate work and recreational activities for most prisoners.5 While many

1 Charles Dickens visited the Cherry Hill, Pennsylvania prison in 1842 and reported: I believe that very few men are capable of estimating the immense amount of torture and agony which this dreadful punishment, prolonged for years, inflicts upon the sufferers . . . there is a depth of terrible endurance in it which none but the sufferers themselves can fathom, and which no man has a right to inflict upon his fellow-creature. I hold this slow and daily tampering with the mysteries of the brain, to be immeasurably worse than any torture of the body . . . . CHARLES DICKENS, AMERICAN NOTES 39 (New York, Harper & Bros. 1842). Danish fairy tale author Hans Christian Andersen reported that a similar Pennsylvania-model prison in Sweden, which used solitary confinement, was “a well-built machine—a nightmare for the spirit.” HANS CHRISTIAN ANDERSEN, PICTURES OF SWEDEN 56 (London, Richard Bentley 1851). And the well- known sociologist Alexis de Tocqueville and his colleague Gustav de Beaumont observed that a similar form of solitary confinement tried in Auburn, New York “proved fatal for the majority of the prisoners. It devours the victim incessantly and unmercifully; it does not reform, it kills. The unfortunate creatures submitted to this experiment wasted away . . . .” TORSTEN ERIKSSON, THE REFORMERS: AN HISTORICAL SURVEY OF PIONEER EXPERIMENTS IN THE TREATMENT OF CRIMINALS 49, 260 nn.9 & 10 (1976) (quoting GUSTAVE DE BEAUMONT & ALEXIS DE TOCQUEVILLE, DU SYSTÈME PÉNITENTIAIRE AUX ÉTATS-UNIS, ET DE SON APPLICATION EN FRANCE 13–14 (Paris, Fournier 1833)). For an alternate English translation, see GUSTAVE DE BEAUMONT & ALEXIS DE TOCQUEVILLE, ON THE PENITENTIARY SYSTEM IN THE UNITED STATES AND ITS APPLICATION IN FRANCE 41 (Francis Lieber trans., S. Ill. Univ. Press 1964) (1833). 2 In re Medley, 134 U.S. 160, 168 (1890). 3 Peter Scharff Smith, The Effects of Solitary Confinement on Prison Inmates: A Brief History and Review of the Literature, 34 CRIME & JUST. 441, 467 (2006). 4 Id. 5 See David A. Ward & Thomas G. Werlich, Alcatraz and Marion: Evaluating Super-Maximum Custody, 5 PUNISHMENT & SOC’Y 53, 55–56 (2003); see also Roy D. King, The Rise and Rise of Supermax: An American Solution in Search of a Problem?, 1 PUNISHMENT & SOC’Y 163, 166 (1999) (explaining that while Alcatraz was a strict institution intended to “break spirits,” the notoriously

161 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W state correctional systems designated certain prisons for the most violent prisoners, rarely did those prisons “operate[] on a total lockdown basis as normal routine.”6 Instead, prisons designated as maximum security “generally allowed movement, inmate interaction, congregate programs, and work opportunities.”7 However, starting in 1972 with the creation of the control unit at the new United States Penitentiary at Marion, and escalating with Marion’s total lockdown and the construction of fifty-seven new super-maximum (supermax) prisons in the 1980s and 1990s,8 the model of incarcerating large numbers of prisoners in near total isolation from each other and the outside world proliferated. By the end of 1998, approximately 20,000 prisoners, or close to 2% of all prisoners serving a year or more in American prisons, were incarcerated in supermax prisons.9 In these supermax facilities, all prisoners were isolated in their cells twenty-three hours per day, with virtually no contact with other prisoners or staff, no programming, no congregate recreation, and no contact visits with family or friends.10 Moreover, the use of solitary units throughout the nation’s prisons dramatically expanded, with prison officials using a myriad of terms such as restrictive housing, disciplinary segregation, administrative segregation, and security housing units to denote the practice of solitary confinement.11 In 2000, the Bureau of Justice Statistics reported that approximately 80,000 people were confined in state or federal segregation units, and the data indicated that between 1995 and 2000, “the growth rate in the number of prisoners housed in segregation

draconian prison lacked many of the features of the modern supermax as prisoners were able to communicate between cells and permitted to recreate and work together). 6 CHASE RIVELAND, SUPERMAX PRISONS: OVERVIEW AND GENERAL CONSIDERATIONS 5 (1999) (explaining the rise of the supermax prison in a report for the Department of Justice). 7 Id. 8 See King, supra note 5, at 167; Daniel P. Mears & Jamie Watson, Towards a Fair and Balanced Assessment of Supermax Prisons, 23 JUST. Q. 232, 232–33 (2006) (noting that as of 2006, there were at least fifty-seven supermax prisons in forty states that housed approximately 20,000 prisoners). 9 King, supra note 5, at 164. 10 Id. at 172; Mears & Watson, supra note 8, at 232, 234, 241. 11 Solitary confinement in the United States has been utilized to discipline prisoners for their misconduct while in prison, as an administrative measure to allegedly prevent future violence by prisoners, and in some cases, such as in certain state death rows, to segregate prisoners because of the crimes they have committed. In other countries, such as the Scandinavian countries, solitary confinement is also used in pretrial detention. See ACLU, A DEATH BEFORE DYING: SOLITARY CONFINEMENT ON DEATH ROW 2, 4 (2013); Jules Lobel & Peter Scharff Smith, Solitary Confinement—From Extreme Isolation to Prison Reform, in SOLITARY CONFINEMENT: EFFECTS, PRACTICES, AND PATHWAYS TOWARD REFORM 1, 3–4 (Jules Lobel & Peter Scharff Smith eds., 2020) [hereinafter SOLITARY CONFINEMENT].

162 115:159 (2020) Mass Solitary and Mass Incarceration far outpaced the growth rate of the overall prison population.”12 In 2014, a report by the Yale Law School Liman Center and the Association of State Correctional Administrators estimated that, as of 2014, approximately 80,000 to 100,000 prisoners in state and federal prisons were in some form of restricted housing, defined as twenty-two to twenty-three hours per day isolated in their cells.13 The reasons for this dramatic rise in the nationwide use of solitary confinement and the development of new supermax prisons have not been explored in depth. In particular, there has been little critical discussion of the rise of mass prolonged solitary as a product of the mass incarceration of the last several decades of the twentieth century.14 The standard, simple explanation for the rebirth of mass solitary in American prisons is that it resulted from the significant rise in prison violence fueled in large part by the emergence of prison gangs, which seemed to leave prison officials with no alternative but to isolate the most dangerous, predatory prisoners.15 From this mainstream perspective, the rise of the supermax is tied to the same forces that brought about mass incarceration in that both the explosion of the prison population and the proliferation of supermax prisons were reactions to the rise in societal

12 COMM’N ON SAFETY & ABUSE IN AM.’S PRISONS, VERA INST. OF JUSTICE, CONFRONTING CONFINEMENT 53 (2006), https://www.vera.org/downloads/Publications/confronting- confinement/legacy_downloads/Confronting_Confinement.pdf [https://perma.cc/8BZT-8M7F]; see also Kevin Johnson, Commission Warns of Harm Isolation Can Do to Prisoners, USA TODAY, June 8, 2006, at 14A (reporting on the Commission’s findings). 13 THE LIMAN PROGRAM, YALE LAW SCH. & ASS’N OF STATE CORR. ADM’RS, TIME-IN-CELL: THE ASCA-LIMAN 2014 NATIONAL SURVEY OF ADMINISTRATIVE SEGREGATION IN PRISON, at i, ii, 14–26 (2015) [hereinafter TIME-IN-CELL], https://law.yale.edu/sites/default/files/area/center/liman/document/time-in-cell_combined_- web_august_2015.pdf [https://perma.cc/45BN-TEAX]; see also ALLEN J. BECK, BUREAU OF JUSTICE STATISTICS, NCJ 249209, SPECIAL REPORT: USE OF RESTRICTIVE HOUSING IN U.S. PRISONS AND JAILS, 2011–12, at 1 (2015) (estimating that “[o]n an average day in 2011–12, up to 4.4% of state and federal inmates and 2.7% of jail inmates were held in administrative segregation or solitary confinement”). 14 Two notable exceptions to the lack of critical academic scholarship exploring the reasons behind the rise of prolonged solitary confinement are Professors Keramet Reiter and Marie Gottschalk. See KERAMET REITER, 23/7: PELICAN BAY PRISON AND THE RISE OF LONG-TERM SOLITARY CONFINEMENT (2016) [hereinafter REITER, 23/7] (exploring the reasons behind the rise of the Pelican Bay Prison Security Housing Unit); Keramet Reiter, The Rise of Supermax Imprisonment in the United States, in SOLITARY CONFINEMENT, supra note 11, at 77, 77 (reviewing the origins of supermax prisons in Arizona and California and the role of litigation in shaping such institutions); Marie Gottschalk, Staying Alive: Reforming Solitary Confinement in U.S. Prisons and Jails, 125 YALE L.J.F. 253 (2016) (discussing the historic proliferation of solitary confinement and its current use). 15 See, e.g., King, supra note 5, at 176 (reporting on a survey of prison administrators that claimed that managing violent prisoners, particularly gang members, was the reason for the development of supermax housing).

163 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W violence. As Congressman Robert Kastenmeier argued in opening the 1985 congressional hearings on the continued lockdown of prisoners in Marion, “[P]rison situations often mirror what is happening in society at-large.”16 The increase in violence in prison settings “is not dissimilar” to the violence taking place in society.17 So too, Norman Carlson, Director of the Federal Bureau of Prisons (BOP) in the 1970s and ’80s, testified before Congress that “[p]risons are microcosms of the larger society,” and that it is necessary to “isolate” those who resort to “violence, threats, and intimidation” from society.18 However, recent critical scholarship has critiqued the mainstream perspective that the rise of mass incarceration in the last quarter of the twentieth century was simply a reaction to increasing crime and violence in American streets. Michelle Alexander, a leading critic of the mainstream narrative, has argued that “mass incarceration in the United States . . . emerged as a stunningly comprehensive and well-disguised system of racialized social control that functions in a manner strikingly similar to Jim Crow.”19 Other critiques have also rejected the standard assertion that mass incarceration was simply a response to increased violence by pointing out the nature of the criminal justice system as a mechanism of social control and articulating other causes for the rise of mass imprisonment.20 While some of these writers do not ignore the clear fact that crime rates substantially rose in the latter part of the twentieth century,21 they explain the rise of mass incarceration as a reaction instead to the rise of the civil rights movement and the societal disruption and tumult of the 1960s and ’70s.22 So too, the

16 Marion Penitentiary—1985: Oversight Hearing Before the Subcomm. on Courts, Civil Liberties, & the Admin. of Justice of the H. Comm. on the Judiciary, 99th Cong. 2 (1985) [hereinafter Oversight Hearing] (opening statement of Rep. Robert Kastenmeier, Chairman, Subcomm. on Courts, Civil Liberties, & the Admin. of Justice). 17 Id. 18 Id. at 149 (statement of Norman Carlson, Director, Federal Bureau of Prisons). 19 MICHELLE ALEXANDER, THE NEW JIM CROW: MASS INCARCERATION IN THE AGE OF COLORBLINDNESS 4 (2010). 20 See generally JOHN F. PFAFF, LOCKED IN: THE TRUE CAUSES OF MASS INCARCERATION AND HOW TO ACHIEVE REAL REFORM (2017) (explaining that prison growth has been driven primarily by increased felony-filing by prosecutors); Loïc Wacquant, From Slavery to Mass Incarceration: Rethinking the ‘Race Question’ in the U.S., 13 NEW LEFT REV. 41 (2002) (identifying the trajectory of racial domination in the United States and the need to bolster an eroding caste cleavage as the main impetus behind expansion of America’s penal system). 21 See, e.g., ALEXANDER, supra note 19, at 41; James Forman, Jr., Racial Critiques of Mass Incarceration: Beyond the New Jim Crow, 87 N.Y.U. L. REV. 21, 35 & nn.41 & 42 (2012) (pointing out that many scholars completely ignore the increasing violence, while some do mention it). 22 See generally TODD R. CLEAR & NATASHA A. FROST, THE PUNISHMENT IMPERATIVE: THE RISE AND FAILURE OF MASS INCARCERATION IN AMERICA (2014) (concluding that the relentless punitive spirit

164 115:159 (2020) Mass Solitary and Mass Incarceration rise of mass solitary cannot simply be explained by an increase of violence in society or prisons. This Essay locates the rise of mass solitary in the 1980s in the context of mass incarceration. It explains the dramatic expansion of the use of solitary confinement and the construction of new supermax prisons as an attempt by prison officials and politicians to maintain control of prisons in the face of increasingly radicalized, rebellious prisoners—often, but not exclusively, African Americans—who had organized protests and disobedient conduct in American prisons from the 1960s to the 1980s. As Professor Judith Resnik has persuasively argued, solitary confinement cannot be viewed in isolation from the panoply of harsh prison policies that characterize modern prison management.23 This Essay expands that perspective to analyze how the rise of solitary was connected to the use of mass incarceration as a form of social control. As society became more violent, so too did many prisons, but to view that violence as the underlying cause of the growth of supermax and other segregated confinement obscures the deeper, underlying causes of the rise of mass solitary. Those causes are linked to the rise of mass incarceration itself. As an initial matter, prison officials responded to the growing political activism of the 1960s and ’70s, often led by radical African-American activists, by developing a mass, often racialized system of control in prisons. This trend in prisons ran parallel to the political use of mass imprisonment as a form of social control in reaction to the political movements and disturbances of that era.24 Second, one aspect of the prison population’s tremendous growth is the criminal justice system’s shift to a preventive ascendant in the United States from the 1960s to the early 2000s operated as the rationale for mass incarceration and, together with the co-alignment of an array of forces, gave rise to a type of social experiment in expanded social control coined “The Punishment Imperative”); ALEXES HARRIS, A POUND OF FLESH: MONETARY SANCTIONS AS PUNISHMENT FOR THE POOR (2016) (analyzing the criminal justice system as a method of social control for managing, punishing, and marginalizing a subset of the U.S. population because of their poverty status); PRISON PROFITEERS: WHO MAKES MONEY FROM MASS INCARCERATION (Tara Herivel & Paul Wright eds., 2007) (identifying numerous esoteric private and public industries and companies whose financial motivations coalesced to help cultivate and sustain mass imprisonment); Forman, Jr., supra note 21 (discussing the history of the New Jim Crow thesis and the author’s common ground and differences with the thesis). 23 Judith Resnik, Not Isolating Isolation, in SOLITARY CONFINEMENT, supra note 11, at 89, 89 (“Solitary confinement is discretely troubling but reflective of the structure of U.S. prisons, which are organized to isolate people in a myriad of ways.”). 24 See, e.g., REITER, 23/7, supra note 14, at 52–58 (discussing the powerful legacy of George Jackson in creating a “genuine fear” among California prison officials “that they were losing control of the prisons”); Resnik, supra note 23, at 90 (“[G]overnment officials used their fears of prisoners’ activism to impose hyper-confinement on hundreds of individuals; targeted were many individuals of the Muslim faith who understood their struggle to be part of an international human rights movement.”).

165 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W model—in other words, a shift from punishing people for crimes they committed to punishing dangerousness, namely, locking people up for long periods of time to incapacitate them from committing crimes in the future and to deter others from committing offenses.25 Similarly, the rise of the supermax and prolonged solitary confinement was in part premised on a shift in the rationale for solitary—from a short-term, discrete punishment for alleged prisoner misbehavior in prison to lengthy, often indeterminate incapacitation and isolation of prisoners who were perceived to be dangerous as a preventive measure.26 Third, mass incarceration accelerated the overcrowding of many state prison systems, often resulting in worsening prison conditions and a subsequent rise of turmoil and violence in prisons.27 These disturbances were then used as a justification for the creation of supermax prisons.28 Fourth, the same law-and-order political ideology based on punitiveness and symbolic toughness led to initiatives from politicians, rather than correctional officials, who sought political gain.29 Uncovering the history and causes of the dramatic rise in supermax prisons and the use of prolonged solitary confinement in the 1980s and ’90s is critical to understanding not only how we got to where we are, but how we can end this cruel and inhumane practice.30 The rationale that prolonged

25 Paul H. Robinson, Commentary, Punishing Dangerousness: Cloaking Preventive Detention as Criminal Justice, 114 HARV. L. REV. 1429, 1429 & n.1 (2001). 26 See, e.g., Jules Lobel, Preventive Detention: Prisoners, Suspected Terrorists and Permanent Emergency, 25 T. JEFFERSON L. REV. 389, 389 (2003). 27 See COMM. ON INT’L HUMAN RIGHTS, THE ASS’N OF THE BAR OF THE CITY OF N.Y., SUPERMAX CONFINEMENT IN U.S. PRISONS 9 (2011) (“Supermax confinement became one method to address the problems resulting from this rapid increase in prison population.”); Craig Haney & Mona Lynch, Regulating Prisons of the Future: A Psychological Analysis of Supermax and Solitary Confinement, 23 N.Y.U. REV. L. & SOC. CHANGE 477, 491 (1997) (“[R]apid expansion of the nation’s prison population . . . has meant that most correctional systems are plagued by extreme overcrowding and the serious management and control problems that go with it. Many prison officials appear convinced that the turmoil brought about by increased population pressures can be managed by segregating and isolating prisoners whom they view as especially troublesome.”); see also Brown v. Plata, 563 U.S. 493, 500–02 (2011) (discussing problems associated with overcrowding). 28 See generally Keramet Reiter, The Path to Pelican Bay: The Origins of the Supermax Prison in the Shadow of the Law, 1982–1989, in CAGING BORDERS AND CARCERAL STATES 303, 303–32 (Robert T. Chase ed., 2019) (examining the development of the supermax prison in California in response to prison violence and the call to address the conditions caused by overcrowding); see also Austin v. Wilkinson, 189 F. Supp. 2d 719, 722–25 (N.D. Ohio 2002) (describing the origins of Ohio supermax sparked by a 1993 riot in Ohio’s maximum-security prison, which did not have sufficient cells to house maximum-security prisoners). 29 RIVELAND, supra note 6, at 5; Leena Kurki & Norval Morris, The Purposes, Practices, and Problems of Supermax Prisons, 28 CRIME & JUST. 385, 390–91 (2001). 30 See generally Sadie Dingfelder, Psychologist Testifies on the Risks of Solitary Confinement, AM. PSYCHOL. ASS’N, Oct. 2012, at 10 (reporting on Dr. Craig Haney’s testimony on solitary confinement’s “grave risk of psychological harm”); Haney & Lynch, supra note 27 (examining the history of solitary

166 115:159 (2020) Mass Solitary and Mass Incarceration solitary as the only way to manage very violent prisoners underlies both society’s and courts’ disposition to allow what seems obviously harmful31— confining a person in a small cell; twenty-three hours per day for years; without any programming or physical contact with spouses, families, or friends; and with little exercise except in small individual cages or rooms. While the past decade has witnessed some reform of mass incarceration and increasing public, judicial, and correctional-official concern that solitary confinement has been overused and should be limited or ended,32 this reform spirit often does not address the problem of the seriously violent prisoner. Some critics of mass incarceration have recognized that we will never end mass incarceration until we face the problem of violence openly and honestly.33 So too, opponents of solitary confinement who have focused on advocating for particularly vulnerable populations34 or prisoners who present no serious security threat must confront the problem of what to do with the very violent prisoner. The Fourth Circuit’s recent decision in Porter v. Clarke illustrates this problem.35 Reflecting the scientific consensus, the Fourth Circuit found that the placement of prisoners on Virginia’s death row in prolonged solitary confinement created a “‘substantial risk’ of serious psychological and emotional harm,” and therefore violated the Eighth Amendment. 36

confinement and its case law in the United States to argue for its regulation); Smith, supra note 3 (reviewing the history and literature on solitary confinement and the development of the supermax prison to demonstrate the substantial effects of long-term isolation); Kirsten Weir, Alone, in ‘The Hole,’ AM. PSYCHOL. ASS’N, May 2012, at 54 (reporting on the psychological effects of long-term solitary confinement). 31 As Judge Richard Posner put it, confinement in the “segregation unit involves considerable isolation, sometimes for protracted periods; and the record shows, what anyway seems pretty obvious, that isolating a human being from other human beings year after year or even month after month can cause substantial psychological damage.” Davenport v. DeRobertis, 844 F.2d 1310, 1313 (7th Cir. 1988). 32 See TIME-IN-CELL, supra note 13, at 3, 4; Lobel & Smith, supra note 11, at 1. 33 DANIELLE SERED, UNTIL WE RECKON: VIOLENCE, MASS INCARCERATION, AND A ROAD TO REPAIR (2019); Michelle Alexander, Opinion, Reckoning with Violence, N.Y. TIMES (Mar. 3, 2019), https://www.nytimes.com/2019/03/03/opinion/violence-criminal-justice.html [https://perma.cc/9C3F- 3RL8]. 34 Particularly vulnerable populations include groups such as prisoners housed in solitary who are mentally ill, juveniles, and pregnant women. See, e.g., Madrid v. Gomez, 889 F. Supp. 1146, 1267 (N.D. Cal. 1995) (holding that seriously mentally ill prisoners could not be held in isolation in the Pelican Bay Security Housing Unit); Anne Teigen & Sarah Brown, Rethinking Solitary Confinement for Juveniles, 24 NCSL, no. 20, 2016, available at https://www.ncsl.org/research/civil-and-criminal-justice/rethinking- solitary-confinement-for-juveniles.aspx [https://perma.cc/8Z7T-BZHT] (reporting that as of that date, “[n]ine states recently passed laws to limit or prohibit using solitary confinement for juvenile offenders”). 35 923 F.3d 348, 353 (4th Cir. 2019) (holding that the imposition of prolonged solitary confinement violates the Eighth Amendment). 36 Id. at 361, 364.

167 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

Nonetheless, the Fourth Circuit found that the district court erred in disregarding the State’s argument that legitimate penological considerations justified the challenged conditions on death row. The court explained that had officials presented such legitimate reasons, which in Porter they did not, similar conditions of solitary could be upheld.37 For the Fourth Circuit, “prison officials tasked with the difficult task of operating a detention center may reasonably determine that prolonged solitary detention of the inmate is necessary to protect the well-being of prison employees, inmates, and the public or to serve some other legitimate penological objective.”38 Some courts have noted that the role “legitimate penological interests” plays in Eighth Amendment litigation has been confusing.39 Where prison officials knowingly deprive a prisoner of basic human needs, such as the need for human contact, no invocation of “legitimate penological reasons” should justify such a practice.40 Nonetheless, Justice Anthony Kennedy undoubtedly expressed a sentiment shared by other judges when he claimed that the issue in a judicial challenge to solitary confinement will be “whether workable alternative systems for long-term confinement exist.”41 Uncovering the history of the rise of the supermax and debunking the myth that the supermax was simply a reaction to a rise of prisoner violence also demonstrate that at critical junctures, alternatives to the supermax did, in fact, exist. Nonetheless, the state and federal governments elected to ignore those alternatives. The recognition that alternatives to prolonged solitary confinement do exist, as states such as Colorado and North Dakota have now concluded,42 provides hope that this inhumane practice can and will be ended. The first Part of this Essay recounts the origins of the supermax prison at Marion Federal Penitentiary in the late 1970s and early 1980s and

37 Id. at 362–63. 38 Id. at 363. In a footnote, the court noted that the dissent’s view that the “opinion could ‘interfer[e]’ with prison officials’ ability to safely confine inmates housed at ‘the federal supermax prisons in Colorado and Illinois’ is without merit” because the majority’s decision permitted correctional officials to argue that the penological interest in protecting against violence would outweigh the serious harm to the prisoners of placing them in solitary confinement. Id. at 363 n.2. 39 Id. at 362 (“Notwithstanding the uncertain role of penological justification in conditions of confinement cases . . . .”); Grenning v. Miller-Stout, 739 F.3d 1235, 1240 (9th Cir. 2014) (“The precise role of legitimate penological interests is not entirely clear in the context of an Eighth Amendment challenge to conditions of confinement.”). 40 Jules Lobel, The Liman Report and Alternatives to Prolonged Solitary Confinement, 125 YALE L.J.F. 238, 240–43 (2016). 41 Davis v. Ayala, 135 S. Ct. 2187, 2210 (2015) (Kennedy, J., concurring). 42 See Rick Raemisch, Colorado Ends Prolonged, Indeterminate Solitary Confinement, in SOLITARY CONFINEMENT, supra note 11, at 311, 311–313; Leann K. Bertsch, Reflections on North Dakota’s Sustained Solitary Confinement Reform, in SOLITARY CONFINEMENT, supra note 11, at 325, 325.

168 115:159 (2020) Mass Solitary and Mass Incarceration demonstrates that the rise of mass solitary was more an official reaction to the need to control politically active and disruptive prisoners than to the violence narrative. The second Part explores prison officials’ need to reassert control over their prisoners and draws the parallels between the rise of both mass incarceration and mass solitary as a racialized mechanism of social control. The third Part introduces the preventive paradigm as a model to control prisoners and demonstrates that the concept of preventing future misconduct fueled both mass incarceration and the modern supermax, resulting in minimizing due process restraints and erroneously isolating thousands of people. Finally, the last Part analyzes the current reform movement and the alternatives that have been proffered and utilized to replace solitary, supermax confinement. The Essay concludes that prolonged solitary confinement can be abolished, and that prison officials have alternatives that can safely manage even very dangerous prisoners.

I. THE VIOLENCE NARRATIVE AND THE RISE OF THE SUPERMAX

A. The Violence Narrative The standard explanation for the rise of the supermax and prolonged solitary confinement in the 1980s and ’90s is that a significant rise in violence in American prisons, particularly fueled by the emergence of violent prison gangs, left prison officials with no alternative but to create supermax prisons that placed the most dangerous, predatory prisoners in high-security isolation. A survey conducted by the National Institute of Corrections in 1997 reported that all but one of the jurisdictions that replied claimed that the development of supermax housing was largely a response to the need for better methods of managing “violent and seriously disruptive inmates.”43 For many of these state administrators, the root of the violence came from the activities of gang members.44 As Human Rights Watch noted in its generally critical 1997 report on the Indiana supermax, The rationale behind supermax facilities and units is rather simple: in an era of rampant violence in many prisons, the segregation of dangerous inmates allows inmates in other facilities to serve their time with less fear of assault; the extreme limitations on inmates’ freedom in such facilities protects both staff

43 NAT’L INST. OF CORR., SUPERMAX HOUSING: A SURVEY OF CURRENT PRACTICE 3 (1997) [hereinafter SUPERMAX HOUSING]. 44 Id.

169 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

and inmates; and the harshness of supermax conditions is believed to deter other prisoners from committing acts that might result in their transfer there.45 So too, the then-Director of the Federal Bureau of Prisons, Norman Carlson, testified before Congress that the institution of a prison-wide lockdown at Marion in the 1980s was “the only answer we have at present” for preventing violence by a small number of individuals in prison, and that the lockdown had successfully reduced violence throughout the federal system.46 The Seventh Circuit’s unanimous decision in Bruscino v. Carlson47 is typical of judicial decisions in the 1980s and ’90s dealing with the rise of mass solitary.48 The court in Bruscino rejected the Marion prisoners’ claims that prolonged isolation resulting from the permanent lockdown violated the Eighth Amendment’s prohibition on cruel and unusual punishment.49 Judge Richard Posner, writing for the court, recognized that conditions involved in the Marion lockdown were “depressing in the extreme,” and for prisoners “[t]o live under such conditions [was] sordid and horrible.”50 Nonetheless, the court held that while the “conditions in Marion deserve careful scrutiny . . . they must be evaluated against the background of an extraordinary history of inmate violence” and that “[t]he defendants placed in the record a remarkable narrative of the violence that led up to the lockdown.”51 After recounting in detail the violence at Marion preceding the lockdown, which included the murder of two guards and a number of inmates, the court noted that “[i]f order could be maintained in Marion without resort to the harsh methods attacked in this lawsuit, the plaintiffs

45 JAMIE FELLNER & JOANNE MARINER, HUMAN RIGHTS WATCH, COLD STORAGE: SUPER- MAXIMUM SECURITY CONFINEMENT IN INDIANA (1997), https://www.hrw.org/legacy/reports/1997/usind/ [https://perma.cc/5MK2-XSMX]. 46 Oversight Hearing, supra note 16, at 142–43 (statement of Norman Carlson, Director, Federal Bureau of Prisons). 47 854 F.2d 162 (7th Cir. 1988). 48 See, e.g., Hoptowit v. Ray, 682 F.2d 1237, 1259 (9th Cir. 1982) (holding that lockdown did not violate Eighth Amendment because it was necessary to contain violence); Madrid v. Gomez, 889 F. Supp. 1146, 1280 (N.D. Cal. 1995) (finding that indeterminate solitary confinement in Pelican Bay SHU does not violate Eighth Amendment for those prisoners who are not seriously mentally ill). 49 854 F.2d at 166. 50 Id. at 164, 166. The magistrate judge and district court also rejected as not credible the testimony of numerous prisoners who testified to beatings and other mistreatment and brutality by correctional officials during the lockdown. Bruscino v. Carlson, 654 F. Supp. 609, 613–14 (S.D. Ill. 1987). The court of appeals affirmed this part of the ruling, noting that evidence of the beatings consisted of testimony by inmates “who frequently lie in prisoner rights’ cases,” and that the court was unwilling in any event to disturb the findings of the magistrate and district court judges. Bruscino, 854 F.2d at 166–67. 51 Id. at 164–65.

170 115:159 (2020) Mass Solitary and Mass Incarceration would have a stronger argument that the methods were indeed cruel and unusual punishments.”52 While the court’s suggestion—whether a prison condition (or set of conditions) constitutes torture versus a permissible reaction to prison violence is dependent on whether there are any alternatives—seems wrong as a matter of law, as a practical matter the potential alternatives to the state’s cruel policies loom large in a judge’s decision on an Eighth Amendment claim.53 Similarly, the Supreme Court in Wilkinson v. Austin held that the process due to prisoners prior to placement in the Ohio supermax had to be measured against “[p]rison security, imperiled by the brutal reality of prison gangs,” which “provides the backdrop of the State’s interest.”54 For Justice Kennedy and the unanimous Court, the brutality of gangs and their uncontrollability by normal means meant that “[p]rolonged confinement in [s]upermax may be the State’s only option for the control of some inmates.”55 The history of the development of control units and supermaxes, however, demonstrates that the impetus for the establishment of these units was often not uncontrollable violence amongst gang members or pathological murderers in prisons. Rather, the crackdown occurred as a response to political activism amongst prisoners who disturbed the normal routine of the prison and threatened the control of prison officials. Even where political activism was accompanied by violence or hostage-taking, prisoner actions were often provoked by ignored grievances and demands to reform prison conditions or end racial discrimination.56 Moreover, prison officials themselves often perpetrated the bulk of the violence as retribution for the prisoners’ activities.57

52 Id. at 165. 53 See Lobel, supra note 40, at 239–40 (discussing Justice Kennedy’s focus on workable alternatives and cases in which courts have affirmed long-term solitary confinement for prisoners deemed particularly violent). 54 545 U.S. 209, 227 (2005). 55 Id. at 229. 56 See generally STAUGHTON LYND, LUCASVILLE: THE UNTOLD STORY OF A PRISON UPRISING (2d ed. 2011) (providing a history of the Ohio Lucasville prison riot as a response to the increasing repression and discrimination by prison officials and their failure to respond to prisoner grievances); HEATHER ANN THOMPSON, BLOOD IN THE WATER: THE ATTICA PRISON UPRISING OF 1971 AND ITS LEGACY (2016) (discussing the violent events at Attica in response to the repression and racial discrimination at the prison which were ignored by prison officials despite prisoner grievances and attempts to address the issues peacefully). 57 See, e.g., THOMPSON, supra note 56 (reviewing the events of Attica in 1971 and its aftermath and demonstrating the disproportionate and unnecessary violence perpetrated by prison officials in ending the prison uprising); see also Tom Puleo, Guards’ Beatings of Inmates After Riots Happens in Many Prisons, HARTFORD COURANT (June 27, 1993), https://www.courant.com/news/connecticut/hc-xpm-1993-06-27- 0000100193-story.html [https://perma.cc/6GGA-BKVP].

171 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

B. Marion and the Creation of the Modern Supermax The creation of the control unit and the institution of the permanent lockdown at Marion are often thought of as the precursor and inspiration of the modern supermax, as state prison officials copied what was then called the “Marion Model.”58 The story of the Marion control unit’s development is illustrative of the supermax as a response mainly to disruptive and rebellious prisoners, not necessarily the most violent. Marion opened in 1963 and was designed to hold 525 “adult male felons who are difficult to manage and control.”59 Marion’s control unit, featuring prolonged solitary confinement, was created in 1972, but not in response to escalating prisoner violence.60 Rather, the first group of prisoners placed in solitary confinement at the control unit were prisoners engaged in a nonviolent work stoppage to protest a guard’s beating of a Mexican prisoner.61 In response to the work stoppage, prison officials created what was at that time termed the control unit, and later designated the “Long-Term Control Unit,” thus coining the term “control unit.”62 The Unit’s origins lay in a behavioral modification program, termed the Control and Rehabilitation Effort (CARE), begun at Marion in 1968 and used on prisoners in solitary confinement.63 The program’s purpose was to bring prisoners under staff control, not only physically but psychologically, and it was a key component in the establishment of the control unit in 1972.64

58 Ward & Werlich, supra note 5, at 59 (“[Bruscino v. Carlson] gave legitimacy to what came to be called the ‘Marion Model’. When state prison wardens visited and observed the unprecedented degree of control the Marion staff had over prisoners, several commented that they ‘had died and gone to heaven.’”). 59 American Prisons in Turmoil (Part 1): Hearings Before the H.R. Select Comm. on Crime, 92d Cong. 277 (1972) (statement of George Pickett, Superintendent, Federal Prison, Marion, Ill.); JESSICA MITFORD, KIND AND USUAL PUNISHMENT: THE PRISON BUSINESS 181 (1973). 60 COMM. TO END THE MARION LOCKDOWN, FROM ALCATRAZ TO MARION TO FLORENCE: CONTROL UNIT PRISONS IN THE UNITED STATES 2 (1992), https://www.freedomarchives.org/Documents/Finder/DOC3_scans/3.alcatraz.marion.florence.1992.htm l [https://perma.cc/SV2W-FR44] [hereinafter FROM ALCATRAZ TO MARION TO FLORENCE]; Bonnie Kerness & Jamie Bissonette Lewey, Race and the Politics of Isolation in U.S. Prisons, 22 ATLANTIC J. COMM. 21, 28 (2014). 61 FROM ALCATRAZ TO MARION TO FLORENCE, supra note 60, at 2. 62 Id. 63 Id. The control unit received its first inmates “with a mission which called for a programme of behaviour modification ‘designed to assist the individual in changing his attitude and behavior.’” King, supra note 5, at 167. 64 See MITFORD, supra note 59, at 134–35; FROM ALCATRAZ TO MARION TO FLORENCE, supra note 60, at 2–3.

172 115:159 (2020) Mass Solitary and Mass Incarceration

As recounted by the Seventh Circuit in a class action lawsuit, the facts of the work stoppage and resulting establishment of the control unit were as follows: Appellants were segregated after a general work stoppage on July 17, 1972. The disruption was in violation of prison rules requiring labor of all able-bodied inmates. To thwart the stoppage, Marion officials first confined the entire prison population to their cells. Most inmates were released six days later, on July 24, after seven inmates suspected to be prominent instigators of the mutiny were relegated to segregation, along with ten supporters insistent upon accompanying them. Work apparently resumed as normal for only a short time thereafter. On the afternoon of July 25, a disturbance again put a halt to regular prison activity. Taking no chances with simply isolating the ringleaders, the Marion administration undertook widespread segregation of inmates suspected of insubordination; approximately eighty-six more prisoners were removed from the general population.65 As the district court found, “In all approximately 103 men were placed in segregation as a result of their participation in the work stoppages.”66 District court Judge James Foreman denied plaintiffs’ motion for injunctive relief, finding that the prison officials had not violated plaintiffs’ due process or Eighth Amendment rights because they had been forced to deal with an “unusual situation” of a major disruption of prison life, and “[p]rompt and effective action . . . was required to restore the prison to normalcy.”67 The court of appeals reversed, finding that these prisoners had been found guilty of disciplinary rule infractions without being accorded due process, and remanded to the district court to determine whether their punishment of long-term, indefinite detention in the control unit was disproportionate and thus violative of the Eighth Amendment.68 As the court of appeals noted, “For a single such event [of misconduct], segregation does not and should not exceed a few months, if that long.”69 On remand, the district court addressed the Eighth Amendment claims of thirty-six prisoners still in the control unit after being accorded hearings which complied with due process requirements, and determined that the defendants had not introduced any evidence that any of these prisoners had committed a serious infraction aside from participating in the 1972 prison

65 Adams v. Carlson, 488 F.2d 619, 621–22 (7th Cir. 1973). 66 Adams v. Carlson, 352 F. Supp. 882, 886–87 (E.D. Ill. 1973). 67 Id. at 885–86. 68 Adams, 488 F.2d at 629, 636. 69 Id. at 628.

173 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W work stoppage.70 The court held that “punishing the Plaintiffs by placing them in confinement under the very restrictive conditions imposed . . . for a period of sixteen months constitutes punishment disproportionate to the various offenses with which these Plaintiffs have been charged and, consequently, is violative of the Eighth Amendment’s prohibition against Cruel and Unusual Punishment.”71 The court ordered their release from the control unit to General Population.72 The use of the Eighth Amendment disproportionality analysis to challenge prisoners’ prolonged solitary confinement was unprecedented and had the potential to significantly limit prison officials’ use of such confinement.73 Thus, in repudiating the prison officials’ rationale, the courts recognized that one of the first large-scale long-term control units was not designed to hold violent prisoners, but rather those prisoners who disrupted prison’s normal routine, even where they did so in a nonviolent manner.74 Regaining control of the prison from disruptive activists—not necessarily curbing violence—was the control unit’s original focus. The judiciary’s rejection of confining prisoners for the extended term in the control unit on the charge of instigating or participating in a labor stoppage helped drive the BOP’s decision to tighten the controls at Marion’s General Population and to change the prison’s purpose into one of segregating dangerous or disruptive prisoners.75 This tightening of controls resulted in numerous prisoners again being sent to the control unit in the mid- 1970s. The only difference was that this time prisoners were not assigned to

70 Adams, 368 F. Supp. at 1053. 71 Id. 72 Id. at 1053–54. 73 Michael Deutsch, The Road from Attica: Mass Incarceration and the Emergence of Control Unit Prisons 10 (unpublished manuscript) (on file with journal); see also Alexander A. Reinert, Eighth Amendment Gaps: Can Conditions of Confinement Litigation Benefit from Proportionality Theory?, 36 FORDHAM URB. L.J. 53, 55–56 (2009) (noting the dissonance between the use of disproportionality Eighth Amendment theory in sentencing and its absence in conditions of confinement jurisprudence and arguing for its potential use in challenging solitary confinement). 74 The purpose of the control unit was officially described by the BOP as “to separate those offenders whose behavior seriously disrupted the orderly operation of an institution from the vast majority of offenders who wish to participate in regular institutional programs.” DAVID A. WARD & ALLEN F. BREED, THE UNITED STATES PENITENTIARY, MARION, ILLINOIS: A REPORT TO THE JUDICIARY COMMITTEE, U.S. HOUSE OF REPRESENTATIVES (1984), reprinted in Oversight Hearing, supra note 16, at 9, 10 (1985) (citing Bureau of Prisons Policy Statement, 5212.1, June 1973). Interestingly, the control unit’s original purpose said nothing about protecting prisoners and staff from violence. 75 Deutsch, supra note 73; Telephone Interview with Michael Deutsch, lead counsel in Adams v. Carlson (Sept. 26, 2019). David Ward and Alan Breed state that the new classification system and the changed purpose of Marion had to do with increasing acts of violence and gang activity throughout the federal system in the late 1970s. WARD & BREED, supra note 74, at 10–11.

174 115:159 (2020) Mass Solitary and Mass Incarceration the control unit for disciplinary infractions, which the courts had rejected, but as a so-called administrative, preventive measure. Marion officials claimed that disruptive prisoners were not being punished for specific acts— for which they would have to be found guilty of misconduct at a disciplinary hearing and subjected to proportionate punishment—but were rather classified to indefinite administrative segregation for being unable “to adjust to an open institutional setting.”76 The officials dubbed the harsh segregation imposed on the prisoners as a “special treatment program,” and the control unit was termed the “Control Unit Treatment Program.”77 Again, the prisoners resorted to court action. In Bono v. Saxbe, the prisoner class alleged substantive and procedural due process violations in their placement in the control unit, as well as an Eighth Amendment challenge to the conditions in the unit, which now included 23.5 hours per day in the cell, no group programming, handcuffing prisoners whenever they left their cells except during showers and recreation, subjecting them to humiliating rectal searches, and denying them any contact visits with family or friends.78 The control unit included cells in one wing with solid steel doors which were termed “boxcar” cells.79 Several prisoners had committed suicide in these harsh, isolating conditions, and prisoners had been confined there because they were activists, critics, jailhouse lawyers, or had influence over other prison activists.80 The named plaintiff, Victor Bono, was “a writer, artist, and well-respected long term prisoner” who posed no threat to other prisoners and was only in the control unit due to the two murders that had landed him in prison years before.81 Judge Foreman generally rejected the plaintiffs’ claims and denied injunctive relief, primarily because he found that use of the control unit to “prevent future disruptions within the institution” was rational and did not constitute punishment.82 Nonetheless, the court found that in some cases, allowing prison officials to place prisoners in the control unit when they engaged in the “disruption of the orderly operation of a prison” let officials silence “prison critics,” “religious leaders,” and “economical and philosophical dissidents.” Oftentimes, “no showing was made as to how

76 Bono v. Saxbe, 450 F. Supp. 934, 941 (E.D. Ill. 1978), supplemented by 462 F. Supp. 146 (E.D. Ill. 1978); see also Deutsch, supra note 73, at 13. 77 Deutsch, supra note 73, at 13. 78 450 F. Supp. at 938–40, 946. 79 Id. at 946. 80 Deutsch, supra note 73, at 13–15. 81 Id. at 15. 82 Bono, 450 F. Supp. at 944.

175 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W these persons disrupted the orderly running of the institution.”83 The court also found that placing prisoners in the control unit solely on the basis of the crime for which they were convicted violated due process because “no reasonable prediction of an inmate’s behavior in the prison could be based on the crime for which he was convicted,” nor could an inmate be “punished” for a specific offense by being placed in the control unit, as bypassing the inmate discipline procedure violated BOP policy.84 The court ordered new hearings for all the prisoners in the control unit.85 It also found that the use of “boxcar” cells, with solid steel fronts—a feature that later was to become standard at supermaxes86—constituted cruel and unusual punishment and enjoined their use.87 The court of appeals affirmed the district court’s decision that placing prisoners in the control unit did not violate the Eighth Amendment, finding that “when the Control Unit is used as a preventive measure, the decision to place a prisoner there is not a violation of substantive due process or of the Eighth Amendment.”88 What clearly motivated both the district court and court of appeals was not merely controlling dangerously violent prisoners, but also that prison officials “have an obligation to society in general to keep prisons operating in an orderly manner, and segregation of those who disrupt these institutions is a reasonable way to meet this obligation.”89 The threat of prison disruption, work stoppages, and other collective action, which the district court in Adams v. Carlton termed akin to “an outright mutiny,”90 was at that point driving the creation of the control unit and the tightening of conditions at Marion generally. In 1979, before major violence broke out at Marion, a Task Force of the Bureau of Prisons recommended that the entire Marion prison be made into a modified control unit.91 Action on that recommendation was deferred, at least in part because the BOP authorities were concerned about its legality and that such action would fare badly in federal court.92 However, in 1979, the BOP did add a new, higher security classification to its system, reserving

83 Id. at 943. 84 Id. 85 Id. 86 See, e.g., Austin v. Wilkinson, 189 F. Supp. 2d 719, 724 (N.D. Ohio 2002) (describing cells at Ohio supermax as having “solid metal door[s]”). 87 Bono, 450 F. Supp. at 937, 947. 88 Bono v. Saxbe, 620 F.2d 609, 612 (7th Cir. 1980) (emphasis omitted). 89 Id. at 616. 90 352 F. Supp. 882, 885–86 (E.D. Ill. 1973). 91 See WARD & BREED, supra note 74, at 11. 92 Telephone Interview with David A. Ward, former Professor, Univ. of Minn. (Oct. 1, 2019).

176 115:159 (2020) Mass Solitary and Mass Incarceration

Level 6 (Marion) for “all violent, assaultive, and . . . disruptive inmates.”93 Prison officials within Marion were also considering a general lockdown before the outbreak of major violence there, and thus implemented stricter controls. Prisoners responded to stricter conditions with several major work stoppages in the early 1980s which eventually led prison officials to close the industrial work program at Marion and transfer the machinery used to another prison.94 In 1983, after several brutal murders and a dramatic increase in violence at Marion, the entire prison was placed on lockdown status, where prisoners were held in solitary confinement.95 The lockdown continued on a permanent basis, and, as recounted earlier, was eventually upheld by the Seventh Circuit Court of Appeals in Bruscino v. Carlson as a rational reaction to the extreme violence at Marion in 1983.96 The Marion experience, which laid the foundation for increased use of prolonged solitary confinement in the 1980s and ’90s, undermines the view that the rise of mass solitary in the form of control units and supermaxes was simply a response to increasing violence in prisons. Instead, the control unit developed in response to political, nonviolent disturbances, such as protests against guard abuses or prisoner grievances. The change in Marion’s character from a general population prison to a solitary confinement unit had been planned even before the spate of violence in 1983 which purportedly made the BOP institute the lockdown, leading to suspicions that, as the American Civil Liberties Union put it in congressional testimony, the Bureau was using the murders at Marion as a “pretext to further change the character of Marion.”97 Moreover, the BOP used Marion to isolate prisoners who had not necessarily proven to be violent in prisons, but who were political radicals or revolutionaries.98 As Marion Warden Ralph Aron testified in

93 WARD & BREED, supra note 74, at 10–11. 94 Id. at 12. 95 Bruscino v. Carlson, 854 F.2d 162, 164 (7th Cir. 1988). 96 Id. 97 Oversight Hearing, supra note 16, at 62–63 (Executive Summary of the ACLU National Prison Project Testimony, June 26, 1985, in the Subcomm. on the Courts, Civil Liberties & the Admin. of Justice). 98 For example, Leonard Peltier, the American Indian Movement leader, Sekou Odinga, member of the Black Liberation Army, Alan Berkman, a former doctor who was a medium-security prisoner with no history of violence in prison but a political radical, Sundiata Acoli, Black Panther/Black Liberation Army member, Ray Levasseur, a white political radical, Puerto Rican Nationalists, and other activists such as Oscar López Rivera, Rafael Cancel Miranda, Kojo Grailing Brown, and Tim Blunk were all transferred to Marion despite having no history of violence in prison. NANCY KURSHAN, OUT OF CONTROL: A FIFTEEN-YEAR BATTLE AGAINST CONTROL UNIT PRISONS, at viii, xi, 39, 133, 183–84 (2013); THE NAT’L COMM. TO FREE PUERTO RICAN POLITICAL PRISONERS & POWS & THE COMM. TO END THE MARION

177 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

1975, “The purpose of the Marion control unit [was] to control revolutionary attitudes in the prison system and the society at large.”99

II. SUPERMAX AND CONTROL OF DISRUPTIVE OR REBELLIOUS PRISONERS The spate of prison work stoppages and other collective protests was not confined to Marion.100 Prisons are a microcosm of society, and the increasingly rebellious civil rights movement, along with other political and social movements of the ’60s and ’70s, spilled over into prisons.101 That era witnessed a wave of prison protests. These were often labor protests, but generally had a “broader vision,” with prisoners protesting “against a host of inequities and dehumanizing aspects of their imprisonment.”102 Those strikes and protests were influenced by civil rights movements, and they helped create a prisoners’ rights movement that continues to this day.103 Prisoners

LOCKDOWN, MASS INCARCERATION AND CONTROL UNITS: CRIME CONTROL OR SOCIAL CONTROL? 6– 16 (1995), https://www.freedomarchives.org/Documents/Finder/DOC3_scans/3.mass.incarceration.control.units.19 95.pdf [https://perma.cc/23A4-BKAR]; FROM ALCATRAZ TO MARION TO FLORENCE, supra note 60, at 2. 99 Steve Whitman, The Marion Penitentiary: It Should Be Opened up, not Locked Down, SOUTHERN ILLINOSIAN, Aug. 7, 1988, at D25; see also Alan Eladio Gómez, Resisting Living Death at Marion Federal Penitentiary, 1972, 96 RADICAL HIST. REV. 58, 61, 80 n.10 (2006) (citing testimony in Bono v. Saxbe, 462 F. Supp. 146 (7th Cir. 1978)); E-mail from Michael Deutsch, lead counsel in Bono v. Saxbe, to author (Oct. 31, 2019) (on file with journal). 100 See Note, Striking the Right Balance: Toward a Better Understanding of Prison Strikes, 132 HARV. L. REV. 1490, 1497–1500 (2019) [hereinafter Striking the Right Balance]; see also Gómez, supra note 99, at 65 (prisoners’ struggle emerged strongest in early 1970s). 101 See Oversight Hearing, supra note 16, at 2 (opening statement of Rep. Robert Kastenmeier, Chairman, H. Subcomm. on Courts, Civil Liberties, & the Admin. of Justice); id. at 149 (statement of Norman Carlson, Director, Federal Bureau of Prisons); see also LARRY SIEGEL & CLEMENS BARTOLLAS, CORRECTIONS TODAY 158–59 (4th ed. 2016) (“Donald Clemmer’s classic study of the prison community at Menard Correctional Center in Illinois . . . notes the existence of numerous parallels between prison and the free world. Clemmer writes, ‘In a sense the prison culture reflects the American culture, for it is a culture within it.’ [But] [s]ome argue that rather than the prison being a microcosm of the larger society, it is a distorted image of that society. In its analysis of the 1971 Attica Prison rebellion, the New York State Special Commission on Attica stated, ‘While it is a microcosm reflecting the forces and emotions of the larger society, the prison actually magnifies and intensifies these forces, because it is so enclosed.’”); Interview by Jonah Walters with Heather Ann Thompson, Professor, Univ. of Mich., available at https://jacobinmag.com/2020/04/prisons-coronavirus-pandemic-heather-ann-thompson/ [https://perma.cc/FD24-FD6W] (“Prisons really are microcosms of the broader society.”). 102 Striking the Right Balance, supra note 100, at 1498–99. 103 See, e.g., ERIC CUMMINS, THE RISE AND FALL OF CALIFORNIA’S RADICAL PRISON MOVEMENT 126–27 (1994) (discussing the organizing of California’s prison movement and the influence of the radical political movement in California); Striking the Right Balance, supra note 100, at 1499–1500.

178 115:159 (2020) Mass Solitary and Mass Incarceration sought in many cases to organize unions.104 The increasing influence of Black Muslim—and later Black Panther—protests were signs of a “wider discontent among inmates.”105 And often, prisoners’ discontent resulted not only in work stoppages but in riots when their grievances were ignored or inadequately addressed. More than 300 prison riots occurred across the United States between 1971 and 1986.106 The establishment of the control unit at Marion followed shortly after the riot and the brutal retaking of the prison at Attica in 1971, as well as revolutionary African-American prisoner George Jackson’s killing by prison guards several weeks earlier in California.107 As Professor Keramet Reiter recounts, prison guards and wardens, such as Carl Larson of California, believed that the civil rights and social justice movement outside of prisons aligned with a Black-led, revolutionary, and violent movement inside prisons. As Larson describes, We had this ‘revolution,’ and it manifested itself with a lot of rhetoric . . . and thought. [But] in the prisons, it manifested in a lot of violence . . . The Black Guerilla Family and the Black Panthers, they had a political side . . . but they were mostly gangs, mafia.108 For Larson and other prison officials, the “national revolutionary movement that culminated with George Jackson” was critical to the understanding of why California built its supermax and the BOP created the Marion Control Unit.109 The “Angola Three” case, where three Black Panther members were placed in solitary for what would be decades based on their political ideology, is another illustration of prison officials reacting to the black revolutionary movement—irrespective of violence—to place radicals and activists in solitary.110

104 CUMMINS, supra note 103, at 252–53 (recounting how in the mid-1970s, the California Prisoners’ Union came close to recognition); see also Jones v. N.C. Prisoners’ Labor Union, Inc., 433 U.S. 119, 121 (1977) (rejecting claims by prisoners to be allowed to organize unions and have union meetings). 105 CUMMINS, supra note 103, at 79; see generally DAN BERGER, CAPTIVE NATION: BLACK PRISON ORGANIZING IN THE CIVIL RIGHTS ERA (2014) (discussing the influence of Black Muslims and Black Panthers in Black prison organizing). 106 REITER, 23/7, supra note 14, at 57; see also BERT USEEM & PETER KIMBALL, STATES OF SIEGE: U.S. PRISON RIOTS, 1971–1986, at 3 (1991). 107 See REITER, 23/7, supra note 14, at 45–51 (recounting the details of Jackson’s alleged attempted escape with a gun and the bloody aftermath that followed). For a thorough review of the Attica events, see THOMPSON, supra note 56. 108 REITER, 23/7, supra note 14, at 40. 109 See id. at 52; see also supra note 99 and accompanying text (discussing view of Marion warden that solitary was designed to curb revolutionary ideas and movements). 110 See ROBERT HILLARY KING, FROM THE BOTTOM OF THE HEAP: THE AUTOBIOGRAPHY OF BLACK PANTHER ROBERT HILLARY KING 21 (2009). See generally ALBERT WOODFOX WITH LESLIE GEORGE,

179 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

That control of collective activity and radical thought has been a primary goal of supermax confinement is illustrated by officials’ often punitive reaction to hunger strikes and other forms of nonviolent protest and statements by prisoners. In 2012, when California Security Housing Unit (SHU) prisoners engaged in nonviolent hunger strikes as what they felt was the only method left to publicize their prolonged solitary confinement after grievances and lawsuits had failed, California prison officials punished them through disciplinary proceedings for participating in or leading the strike. 111 In 2012, prisoner representatives from different ethnic groups signed on to an Agreement to End Hostilities between the different racial groups that make up the California prison system in a collective effort to end the racial violence that had beset the California prisons for decades.112 In response, California officials stated in the Ashker v. Brown litigation that the Agreement was evidence of the continuing threat to prison security posed by the plaintiffs in that it showed the “influence” they had over other prisoners in the prisons.113 Breaking the control and influence of these leaders was paramount to prison officials, even though the leaders’ influence was being used to put an end to racial violence in California prisons.114 Moreover, prison officials were beset by increasing litigation, both from civil rights lawyers and from prisoners themselves, which began to place significant restrictions on what had been the virtually absolute discretion of officials to manage prisoners.115 Just as civil rights protests and

SOLITARY (2019) (recounting his experience serving four decades in solitary confinement as one of the Angola 3); ANGOLA 3: BLACK PANTHERS AND THE LAST SLAVE PLANTATION (Obstacle Illusions 2008) (recounting the story of the Angola 3 and their record stay in solitary confinement); IN THE LAND OF THE FREE . . . (The Mob Film Company 2010) (same). 111 Plaintiffs’ Second Amended Complaint at ⁋⁋ 157, 164, Ashker v. Brown, No. 4:09-cv-05796-CW (N.D. Cal. Sept. 10, 2012). 112 PELICAN BAY STATE PRISON-SECURITY HOUSING UNIT SHORT CORRIDOR HUNGER STRIKE REPRESENTATIVES, AGREEMENT TO END HOSTILITIES (2012), https://ccrjustice.org/sites/default/files/attach/2015/07/Agreement%20to%20End%20Hostilities.pdf [https://perma.cc/7H6U-BNG5] [hereinafter AGREEMENT TO END HOSTILITIES]. 113 Defendants’ Notice of Motion & Motion to Dismiss Plaintiffs’ Second Amended Complaint, or Alternatively, to Stay; Memorandum of Points and Authorities at 2 & n.1, Ruiz v. Brown, No. 4:09-cv- 05796-CW (N.D. Cal. Dec. 17, 2012). 114 While it is possible to argue that the Agreement did demonstrate the influence that the leaders of the hunger strikes had in the California prisons and that such influence could be used for negative purposes, the point is that the Agreement had as its objective an incredibly positive goal—namely ending the ethnic violence in California prisons. California ignored that goal because it was so focused on breaking these leaders’ influence over other prisoners. See AGREEMENT TO END HOSTILITIES, supra note 112. 115 See CUMMINS, supra note 103, at 80–82 (explaining how the enormous flood of habeas petitions from California state prisoners left prison officials unsure how to respond); MALCOLM M. FEELEY & EDWARD L. RUBIN, JUDICIAL POLICY MAKING AND THE MODERN STATE: HOW THE COURTS REFORMED

180 115:159 (2020) Mass Solitary and Mass Incarceration litigation were legally dismantling Jim Crow, the combination of prison protests and litigation was eliminating the “Plantation Model” which had been ensconced in southern prisons until the 1970s, and more generally limiting the ability officials had to punish and control prisoners.116 An increase in collective prison action in the 1970s and 1980s, combined with an increase in successful prisoner litigation, determinate sentencing reforms, and the rise of prison gangs during this period led officials to conclude their prisons were out of control and they had lost their traditional management tools.117 As Professor Reiter points out in discussing California prisons: “California prison officials seemed to be losing both their autonomy to run their prisons free from public scrutiny and their discretion over the intensity and severity of prisoners’ punishments.”118 Both officials and politicians responded by proposing and designing harsher forms of managing the prisoners they perceived to be most dangerous to the prison order—the “George Jacksons.”119 As early as 1972, at around the same time the Marion Control Unit was established, California Governor Ronald Reagan called for the development of new, high-tech maximum-security prisons to incarcerate troublemaking convicts.120 The rise of mass solitary confinement thus springs from the same root cause that critical theorists identify as inspiring mass incarceration: the need to develop new mechanisms of social control to replace an old order thrown into turmoil by mass protests, litigation, and changing societal attitudes.121 In

AMERICA’S PRISONS 30–51 (1998) (describing first what the authors term the “hands-off era” of federal court responses to prisoner complaints which accorded prison officials virtually complete discretion over running prisons and then recounting the dramatic effect of prison litigation). 116 See generally Malcom M. Feeley & Van Swearingen, The Prison Conditions Cases and the Bureaucratization of American Corrections: Influences, Impacts and Implications, 24 PACE L. REV. 433 (2004) (exploring why judicial intervention has been successful in dismantling the “Plantation Model” and other forms of maltreatment against prisoners); Whitney Benns, American Slavery, Reinvented, ATLANTIC (Sept. 21, 2015), https://www.theatlantic.com/business/archive/2015/09/prison-labor-in- america/406177/ [https://perma.cc/2ZGH-RFWZ] (explaining that at Angola, where inmates can be forced to work without compensation under threat of punishment as severe as solitary confinement, slavery never ended but, in fact, was merely reinvented). 117 REITER, 23/7, supra note 14, at 84. 118 Id. at 78. 119 Id. at 84; see also CUMMINS, supra note 103, at 248. 120 CUMMINS, supra note 103, at 248 & n.94. 121 ALEXANDER, supra note 19, at 13 (arguing that mass incarceration developed as a reaction to the demise of Jim Crow as a mechanism of social control and represented a new form of social control); Wacquant, supra note 20, at 52 (explaining that mass incarceration represented the backlash against the advances won by the social movements and “offered itself as the universal and simplex solution to all manners of social problems,” particularly the violent urban upheavals of the mid-’60s; “[a]s the walls of the ghetto shook and threatened to crumble, the walls of the prison were correspondingly extended, enlarged and fortified.”).

181 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W both cases of mass isolation and removal from society, the political technique involved the imagery of a violent, predatory monster who was no longer perceived to be human. The supermax and the rise of mass prolonged solitary confinement “represent the application of sophisticated, modern technology dedicated entirely to the task of social control.”122 Moreover, as with mass incarceration, the supermax represents a form of control different from, yet connected to, the racist practices used to brutalize, control, and subordinate African Americans in the plantation system and convict labor system of previous eras. As Professor Angela Davis argues: The ultimate manifestation of this phenomenon [of racism in the prison system] can be found in the supermax prison, whose main function is to subdue and control “problematic” imprisoned populations—again, composed largely of black men—who, having been locked away in the most remote and invisible of spaces, basically are no longer thought of as human.123 Mass solitary functions as a form of social control against disobedient, disruptive, rebellious, or violent prisoners using three main mechanisms. The first, and most obvious, is long-term, total physical isolation. The modern supermax uses sophisticated technology to ensure minimal contact between staff and prisoners combined with maximum and usually remote surveillance of each prisoner.124 This minimizes potential for either individual attacks on staff or other prisoners or collective action by prisoners.125 Second, and equally important, is the element of psychological control, effectuated by not only the physical isolation, but the various psychological elements used to attempt to break the spirit and resistance of the prisoner. Justices John Paul Stevens and Ruth Bader Ginsburg described the deprivation of reading materials and personal photographs for particularly disruptive prisoners at the Pennsylvania supermax, which the majority of the

122 Craig Haney, “Infamous Punishment”: The Psychological Consequences of Isolation, NAT’L PRISON PROJECT J. 3, 3 (1993). 123 Angela Y. Davis, Race, Gender, and Prison History: From the Convict Lease System to the Supermax Prison, in PRISON MASCULINITIES 35, 44 (Don Sabo, Terry A. Kupers, & Willie London eds., 2001). 124 See Haney, supra note 122, at 3; see generally Charles A. Pettigrew, Technology and the Eighth Amendment: The Problem of Supermax Prisons, 4 N.C. J.L. & TECH. 191, 194–98 (2002) (discussing how “[a]dvanced technology distinguishes Supermax prisons from their conventional counterparts and allows for the isolation of prisoners”). 125 Pettigrew, supra note 124, at 195. However, this potential is not entirely eliminated, as the various hunger strikes by prisoners at California’s Pelican Bay Security Housing Unit (SHU) illustrate.

182 115:159 (2020) Mass Solitary and Mass Incarceration court upheld as a rational behavioral-management effort, as coming “perilously close to a state-sponsored effort at mind control.”126 California’s policy to break prisoners placed in the SHU for alleged gang affiliation was to condition release from isolation on the prisoner becoming an informant.127 In California, the attempt to break the prisoner through informing (known as “debriefing”) included conditioning virtually any human contact on debriefing.128 In one such instance, a prisoner was told after his parent died that the only way he could receive any additional phone calls to his family was by becoming an informant.129 Behavioral management control is ubiquitous in supermax prisons and control units, another legacy of the Marion experience. An article published in Corrections Management Quarterly tellingly concluded that “in control units[,] mind control is a primary weapon.”130 As John McCain reflected on his experience of solitary confinement as a prisoner of war: “It’s an awful thing, solitary. It crushes your spirit and weakens your resistance more effectively than any other form of mistreatment.”131 Finally, the supermax, and officials’ invocation of the need to protect inmates and staff from the dangerous, pathological predator, allowed prison officials to recover in certain respects the near absolute discretion and authority of a bygone era.132 While the first legal challenges of the early

126 Beard v. Banks, 548 U.S. 521, 552 (2006) (Stevens, J., dissenting). Contra id. at 531 (majority opinion). 127 REITER, 23/7, supra note 14, at 145–46 (revealing through personal narratives of prisoners that the only ways to get out of the SHU are “parole, snitch, or die”). 128 Griffin v. Gomez, 741 F.3d 10, 13–14 (9th Cir. 2014) (recounting the district court opinion holding that keeping Griffin in isolation in Pelican Bay SHU for twenty years with his only way out being becoming an informant and thereby risking his and his family’s safety violated the Eighth Amendment); Plaintiffs’ Second Amended Complaint, supra note 111, at ⁋ 7 (“Plaintiffs’ and class members’ only way out of isolation is to ‘debrief’ to prison administrators (i.e., report on the gang activity of other prisoners) . . . . Accordingly, for those many prisoners who refuse or are unable to debrief, defendants’ policies result in ‘effectively permanent’ solitary confinement.”); REITER, 23/7, supra note 11, at 145–46 (particularly for inmates with life sentences, “[s]nitching was the only way [prisoners] could expect to leave the SHU alive”). 129 See Plaintiffs’ Second Amended Complaint, supra note 111, at ⁋ 52 (recounting that after a prisoner’s mother died he was allowed to make a phone call, which was the only call to family or friends that he had been allowed in nine years, but was immediately told by prison officials “to think about taking advantage of the debriefing program”). 130 Rodney J. Henningsen, W. Wesley Johnson & Terry Wells, Supermax Prisons: Panacea or Desperation?, 3 CORRECTIONS MGMT. Q. 53, 58 (1999). 131 JOHN MCCAIN & MARK SALTER, FAITH OF MY FATHERS 206 (1999). 132 See REITER, 23/7, supra note 14, at 5, 7 (“Sociolegal scholars call the resulting pattern of compliance ‘legal endogeneity’: courts impose minimum standards of humane treatment, prison officials redefine minimum standards as establishing prisoners’ maximum privileges, and the courts defer. . . .

183 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

1970s imposed some significant restraints on official discretion,133 by the mid- to late 1990s, the courts had developed a largely hands-off policy on administrative segregation and supermax confinement.134 Even the Supreme Court’s decision in Wilkinson v. Austin in 2005 recognized that prisoners had a liberty interest in avoiding prolonged solitary confinement in the Ohio supermax, although the Court only required prison officials to follow minimal due process procedures.135 The upshot of the judiciary’s legitimation of supermax confinement was that for decades officials had almost absolute control over who could be placed in the supermax and for how long and, short of physical brutality, the treatment they received once so placed. As Professor Davis has argued, this absolute authority is reminiscent of the impunity with which the convict lease system operated in total disregard of the humanity of the mostly Black prisoners.136 In a sense, the supermax perfected the disciplinary, social control functions of the modern prison first articulated by the brilliant French philosopher Michel Foucault.137 Foucault explains the modern prison as a panoptic method of control, in which the architecture of the prison is adopted for constant surveillance, which along with branding the individual as “dangerous/harmless” or “normal/abnormal,” permits the state to engage in “coercive assignment” and control and regulate behavior.138 Interestingly,

[W]hen officials said the prisoners in Pelican Bay were uniformly dangerous and that the [supermax] was absolutely necessary, judges took them at their word.”). 133 See supra notes 115–116 and the prison litigation described in REITER, 23/7, supra note 14, at 68–70. 134 See In re Long Term Admin. Segregation of Inmates Designated as Five Percenters, 174 F.3d 464 (4th Cir. 1999), abrogation recognized by Latson v. Clarke, 794 F. App’x. 266, 270 (4th Cir. 2019); Bruscino v. Carlson, 854 F.2d 162, 168 (7th Cir. 1988); see also Sandin v. Conner, 515 U.S. 472, 484– 86 (1995) (explaining that thirty days in solitary confinement did not constitute an “atypical and significant hardship” giving rise to a liberty interest requiring prison officials to accord prisoners any due process prior to placement). 135 545 U.S. 209, 229–30 (2005). 136 Davis, supra note 123, at 44; see also Convict Leasing, EQUAL JUST. INITIATIVE (Nov. 1, 2013), https://eji.org/news/history-racial-injustice-convict-leasing/ [https://perma.cc/L5YV-TGHT] (“After the Civil War, slavery persisted in the form of convict leasing, a system in which Southern states leased prisoners to private railways, mines, and large plantations. While states profited, prisoners earned no pay and faced inhumane, dangerous, and often deadly work conditions. Thousands of black people were forced into what authors have termed ‘slavery by another name’ until the 1930s.”). Convict leasing was thus clearly related to the plantation model of prisons. Benns, supra note 116. 137 See MICHEL FOUCAULT, DISCIPLINE AND PUNISH: THE BIRTH OF THE PRISON 201–02 (Alan Sheridan trans., 1977). 138 Id. at 199–210; see also Sandra McGunigall-Smith & Robert Johnson, Escape from Death Row: A Study of “Tripping” as an Individual Adjustment Strategy Among Death Row Prisoners, 6 PIERCE L. REV. 533, 543 (2008) (“SHUs are the epitome of the panoptic gaze that lay at the heart of the disciplinary society envisioned by Foucault.”).

184 115:159 (2020) Mass Solitary and Mass Incarceration given today’s coronavirus context, Foucault traces the origins of the modern prison system to towns consumed by medieval plague, where the people were isolated from each other and subjected to constant surveillance. 139 He also notes the turn to solitary confinement in French history, following a period of political agitation and revolt: “The wave of revolt . . . and perhaps the general agitation in the country in the years 1842–3 resulted in the adoption in 1844 of the Pennsylvanian régime of absolute isolation . . . .”140 In the United States, the rise of mass solitary, as with the growth of mass incarceration, was racially discriminatory. Various studies indicate that the racial disparities that characterize the prison population generally, as compared to the United States population as a whole, are replicated amongst those placed in solitary confinement, although not in as extreme a form. A 1980 statistical analysis of a single medium-security prison in the South demonstrated, albeit on a small scale, “that the race of an inmate was correlated with the disciplinary decisions of correctional officers.”141 The study found that “black and white inmates were equally likely to engage in rule-breaking activity,” yet “they were not equally likely to be reported for rule infractions.”142 Similarly, when Wisconsin opened its supermax prison, of its first 215 inmates, approximately 60% were African-American, with Hispanics constituting almost all of the rest.143 These statistics are startling in a state where 46% of the prison inmates were African-American and 17% were Hispanic, and yet only 5% of the state’s population was African- American and less than 2% was Hispanic.144 More recently, the 2014 ASCA-Liman Report found that in the twenty- two reporting jurisdictions, African-American males constituted 48% of the

139 Foucault writes that in a medieval town consumed by the plague: Each individual is fixed in his place. And, if he moves, he does so at the risk of his life, contagion or punishment. Inspection functions ceaselessly. The gaze is alert everywhere: ‘A considerable body of militia, commanded by good officers and men of substance’, guards at the gates, at the town hall and in every quarter to ensure the prompt obedience of the people and the most absolute authority of the magistrates, ‘as also to observe all disorder, theft and extortion’. At each of the town gates there will be an observation post; at the end of each street sentinels. FOUCAULT, supra note 137, at 195–96. 140 Id. at 318 n.6. That adoption was short-lived and repealed in 1847. Id. 141 Andrea C. Armstrong, Race, Prison Discipline, and the Law, 5 U.C. IRVINE L. REV. 759, 761 (2015). 142 Id. at 765 (quoting Eric D. Poole & Robert M. Regoli, Race, Institutional Rule Breaking, and Disciplinary Response: A Study of Discretionary Decision Making in Prison, 14 LAW & SOC’Y REV. 931, 944 (1980)). 143 Jerry R. DeMaio, Comment, If You Build It, They Will Come: The Threat of Overclassification in Wisconsin’s Supermax Prison, 2001 WIS. L. REV. 207, 229. 144 Id. at 229 n.129.

185 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W male population in administrative segregation, as compared to 39% of the general prison population.145 The reported demographics for female prisoners demonstrated an even more significant racial disparity, with Black female prisoners constituting only 23% of the total female custodial population but 35% of the female restricted housing population across the jurisdictions reporting.146 Those statistics do not include California, which did not participate in the ASCA-Liman study,147 and has an extremely disproportionate racial/ethnic balance in its usage of solitary. In California, “Latinos made up 42 percent of the general prison population, but 86 percent of those in solitary confinement. Whites, by contrast, were 22 percent of the general population, but only nine percent of those in solitary.”148 In 2016, the New York Times published a comprehensive report finding that “racial disparities were embedded in the [state] prison experience in New York.”149 According to the report, Black and Latino prisoners were disciplined at higher rates than white prisoners—in some cases twice as often—and were sent to solitary confinement more frequently and for longer durations than their white counterparts.150 “At Clinton, a prison near the Canadian border where only one of the 998 guards is African-American, black inmates were nearly four times as likely to be sent to isolation as whites, and they were held there for an average of 125 days, compared to 90 days for whites.”151 The disparities in disciplinary sanctions were often greatest for “vaguely defined” infractions that gave discretion to officers and did not require production of physical evidence, like disobeying a direct order.152 Indeed, vaguely defined infractions have been a powerful tool in perpetuating supermax facilities and prolonged solitary confinement. In a state where Blacks make up only 14% of the general population but nearly half of the prison population,153 these reports indicate that the racial character of mass incarceration is repeated in the use of mass solitary confinement. As

145 TIME-IN-CELL, supra note 13, at 31. 146 Id. at 36. 147 Id. at 76 n.101. 148 Juleyka Lantigua-Williams, The Link Between Race and Solitary Confinement, ATLANTIC (Dec. 5, 2016), https://www.theatlantic.com/politics/archive/2016/12/race-solitary-confinement/509456/ [https://perma.cc/LAQ8-ZHWF]. 149 Michael Schwirtz, Michael Winerip & Robert Gebeloff, The Scourge of Racial Bias in New York State’s Prisons, N.Y. TIMES (Dec. 3, 2016), https://www.nytimes.com/2016/12/03/nyregion/new-york- state-prisons-inmates-racial-bias.html [https://perma.cc/7EE3-VQYA]. This report was based on tens of thousands of disciplinary cases against New York State prisoners. Id. 150 Id. 151 Id. (emphasis added). 152 Id. 153 Id.

186 115:159 (2020) Mass Solitary and Mass Incarceration with mass incarceration, mass solitary constitutes a racialized system of control.154 The problem of how to reassert social and political control in the wake of civil rights protests, urban rebellions and prison work stoppages, protests, and riots of the ’60s and ’70s drove both political leaders and prison officials to develop new mechanisms of control. By utilizing the narrative of the increasing violence and disruption in American society and prisons, officials were able to institute both the massive increase of incarceration and the rise of the supermax within prisons to isolate people from society and humanity in an attempt to end racial and political protest and change.

III. THE PREVENTIVE MODEL The Marion experience also illustrates the shift from the disciplinary model of placing people in solitary for determinate periods of time as punishment for specific misconduct, utilized generally throughout the twentieth century, to the preventive model of the modern supermax, where prisoners are held in solitary indefinitely as a measure to prevent future violence or disturbances. The preventive paradigm utilizes exceedingly vague criteria containing no clear indication to prisoners as to what conduct will result in their placement in solitary and how they can eventually earn their release.155 Prison officials argue that because solitary confinement is not “punitive” but instead “preventive,” prisoners are afforded less rights and the actions of officials are to be accorded less scrutiny by the courts.156 This

154 Professor Andrea Armstrong writes, “[M]inority offenders may be more likely to be perceived as a disciplinary threat by correctional officers, regardless of an offender’s actual behavior.” Armstrong, supra note 141, at 770 (emphasis omitted). Moreover, the implicit biases of prison guards is particularly relevant in cases of “minor or ambiguous conduct charges” where “‘vaguely worded “catchall” rules’ . . . almost always pertain to an inmate’s attitude rather than conduct” because these rules are “especially susceptible to . . . influence by an individual prison guard’s implicit racial preferences.” Id. at 770–72 (emphases omitted). 155 See Austin v. Wilkinson, 189 F. Supp. 2d 719, 731 n.14 (N.D. Ohio 2002) (“The memorandum lists the following behavior as criteria for classification to high maximum security status: The inmate’s conduct or continued presence at the sending institution poses a serious threat to . . . the security of the prison; The nature of the inmate’s criminal offense indicates that the inmate poses a serious threat to the physical safety of any person, or to the security of the prison . . . .”); Johnson v. Wetzel, 209 F. Supp. 3d 766, 771–72 (M.D. Pa. 2016) (stating that a prisoner was retained in indefinite solitary confinement for a period of thirty-six years despite having committed no serious misconduct for the last twenty-eight years, on the basis of his escape history and undefined “threats to harm others”). 156 Ohio argued in the Supreme Court case of Wilkinson v. Austin that because their decisions in placing prisoners in the supermax were predictive of dangerousness, namely preventive, they did not have to accord the prisoners the full due process protections provided for disciplinary proceedings. Brief for Petitioners at 19–20, Wilkinson v. Austin, 545 U.S. 209, 224–29 (2005) (No. 04-495) (“[W]here the

187 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W paradigm allows prison officials wide discretion to assert control over certain allegedly disruptive prisoners and the ability to threaten the rest of the prison population with similar consequences if they misbehave or associate with the wrong people. The ascent of the preventive paradigm for prolonged solitary confinement is thus tied to the rise of the “violent prisoner” narrative and the use of the supermax to reassert officials’ control.157 Judicial decisions affirm the dichotomy between disciplinary and preventive detention, as the Supreme Court’s opinion in Wilkinson v. Austin demonstrates. In Wilkinson, the Court upheld a lower court’s finding that prisoners had a liberty interest in avoiding placement in Ohio State Penitentiary (OSP), the Ohio supermax. However, the Court determined that the due process requirements for prisoners facing indeterminate solitary confinement for preventive reasons are minimal.158 The upshot of that decision was that a prisoner in Ohio accused of murdering another prisoner, for example, could be disciplined and sent to segregation for a determinate term, after a due process hearing which comported with the protections set forth by the Court in Wolff v. McDonnell.159 Yet, officials could also send a prisoner to supermax for years for relatively minor offenses while affording them less and weaker due process protections. For instance, one of the named plaintiffs in Wilkinson, Daryl Heard, was charged with conspiring to bring marijuana into the prison and “punished” by receiving fifteen days in disciplinary segregation. He then was transferred from a medium-security prison to the supermax as a “preventive” administrative measure for drug distribution, with only minimal due process, and kept in the supermax for years.160 Indeed the district court in Wilkinson found that more than fifty people in Ohio were transferred to the supermax even though they had committed no violence in prison and their only rules violation was drug involvement while in prison.161 Like mass incarceration, the use of solitary confinement as a preventive measure is often justified by the desire to limit gang violence. While gangs undoubtedly present serious problems in prisons, the preventive paradigm is problematic because it allows prison

decision is predictive, the other Mathews factors, especially the private interest affected [namely, the prisoner’s interest], carry very little weight, if any.”). 157 See supra Part I (discussing violent prisoner narrative) and Part II (discussing supermaxes and prison officials reasserting their control). 158 Wilkinson v. Austin, 545 U.S. 209, 224–29 (2005). 159 418 U.S. 539 (1974). 160 Austin, 189 F. Supp. 2d at 730, 731 n.14. 161 Id. at 736.

188 115:159 (2020) Mass Solitary and Mass Incarceration administrators broad discretion to send and confine prisoners to supermax facilities.162 The preventive paradigm for supermax confinement reached its apex in California’s Security Housing Unit (SHU) at Pelican Bay State Prison. There, any California prisoner allegedly affiliated with a prison gang could be sent to solitary confinement based simply on their association, without any evidence or act of prison misconduct.163 In 2011, more than 1,000 prisoners were assigned to preventive, so-called “administrative” detention, and held in very isolating conditions. Prisoners were confined to small, eighty-square-foot windowless cells for 22.5 to 24 hours per day.164 Phone calls with family or friends were prohibited, as were contact visits with any visitors. Prisoners left their cells only for approximately ninety minutes per day to recreate alone in a facility with high walls and a partial grate covering the top so that they received virtually no direct sunlight. The recreation area, only several times larger than their cells, was devoid of anything but one handball. While prisoners were able to communicate with each other by shouting through the walls, social contact was limited, disembodied, and sometimes punished. The prisoners had virtually no educational or other programming, and no work or vocational programs. These prisoners had not seen trees, birds, or grass for years, had not touched another human being

162 Id. at 748–49, 751 (“The Department chose to move Roe to a more secure and more expensive facility without articulating a single affirmative action he had undertaken. Instead, he was allegedly moved because of longtime gang membership and his involvement in a racial incident more than five years ago. These justifications ring hollow . . . . [Similarly,] Thompson had a classification review. The hearing committee’s worksheet indicated that Thompson had no rules violation findings, no administrative control placement, and no violence in the last five years . . . . Nonetheless, without any notice of the evidence claimed against him, the Department sent Thompson to the OSP” and the reclassification committee’s recommendations to reduce Thompson’s security classification “were denied because of Thompson’s alleged gang membership”). Under Ohio’s New Policy, which was a reform measure, a prisoner could be sent to the supermax if he was “identified by the institution Security Threat Group Coordinator as a leader, enforcer, or recruiter of a security threat group, which is actively involved in violent or disruptive behavior.” Id. at 751. As the district court pointed out, this provision did not cabin the Coordinator’s discretion—so long as he or she determined that the inmate was a gang leader, the prisoner could be sent to the supermax. Id. 163 CAL. CODE. REGS. tit. 15, § 3378(c)(4) (2013) (“An associate is an inmate/parolee or any person who is involved periodically or regularly with members or associates of a gang. This identification requires at least three (3) independent source items of documentation indicative of association with validated gang members or associates. Validation of an inmate/parolee or any person as an associate of a prison gang shall require at least one (1) source item be a direct link to a current or former validated member or associate of the gang, or to an inmate/parolee or any person who is validated by the department within six (6) months of the established or estimated date of activity identified in the evidence considered.”) (repealed 2014). 164 Plaintiffs’ Second Amended Complaint, supra note 111. Prisoners were only entitled to a fifteen- minute family call in event of an emergency, such as if a close family member died. The amended complaint contains the facts asserted in the rest of this paragraph. See also REITER, 23/7, supra note 14.

189 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W during their time in the SHU, and many had no visitors due to the isolated location of the prison. By 2011, approximately five hundred prisoners at Pelican Bay SHU had been in solitary confinement for over ten years, seventy-eight of them for more than two decades.165 For most, there was no way out. They had not been placed in solitary confinement because of some serious misconduct that they had committed in prison, nor because of the heinousness of their criminal offense. Rather, they had been placed in solitary because of an alleged affiliation or association with a prison gang. They need not be an actual or alleged gang member; the alleged affiliation need not even rise to the level of a gang member, and many of the prisoners confined in the Pelican Bay SHU were classified as associates.166 A prisoner could be labeled a so-called “associate”—defined as someone who is not necessarily a member of the gang but who periodically associates with gang members—and placed in the SHU.167 Tattoos, artwork, political writings, and greeting cards which allegedly had some indicia of gang involvement all sufficed for SHU placement.168 Moreover, prisoners were not put in solitary for a determinate term, but rather were housed there indefinitely. Only once every six years was their placement reviewed, and virtually all were perfunctorily retained

165 Plaintiffs’ Second Amended Complaint, supra note 111, ¶ 33. 166 CAL. CODE REGS. tit. 15, § 3378 (c)(4) (defining an associate of a gang eligible to be sent to the Security Housing Unit (SHU) for an indefinite term as “an inmate/parolee or any person who is involved periodically or regularly with members or associates of a gang”). Under this vague definition almost any prisoner who had some vague and periodic association with a gang could and often would be sent to the SHU. See infra note 167 and accompanying text. 167 CAL. CODE REGS. tit. 15, § 3378 (c)(4). 168 Prisoners were both placed in SHU initially and then retained there as “active” gang affiliates based on indicia of so-called gang association such as tattoos, political writings, and artwork. See Plaintiffs’ Second Amended Complaint, supra note 111, ¶¶ 93, 104, 105. George Ruiz, a sixty-nine-year- old prisoner, spent twenty-eight years in solitary confinement—twenty-two at the Pelican Bay SHU— and was denied inactive gang status for his possession of photocopied drawings alleged to contain symbols associated with a gang. See id. ¶¶ 14, 104. Gabriel Reyes, forty-six, spent fourteen years in isolation at the Pelican Bay SHU and was also denied inactive status on the basis of a tattoo drawing found in his cell which included a geometric pattern (known as the G-Shield) that had been “rejected as a gang-related source item in 1996, 2003, and 2005.” See id. ¶¶ 18, 105. Jeffrey Franklin, fifty-two, spent twenty-two years in Pelican Bay’s SHU and was denied inactive status after his name appeared on another prisoner’s gang roster. He was said to be “communicat[ing] by talking” with a validated member of a different gang thereafter, which was instructed to be considered at his next review. See id. at 15, 106.

190 115:159 (2020) Mass Solitary and Mass Incarceration in solitary.169 The only practical ways out were to be released from prison, to become an informant, or to die.170 Placement and retention at Pelican Bay built on the paradigm established at Marion in the 1970s, where vague standards for who could be placed in the prison’s control unit—coupled with new guidelines as to who could be transferred to Marion’s high security General Population—allowed for a proliferation of arbitrariness in assignment to solitary confinement. 171 Notably, a report investigating the Marion Lockdown concluded that 80% of inmates at Marion in the fall of 1984 had security ratings that would have normally required placement at a lower-security institution, not a Level 6 prison, such as Marion.172 Nonetheless, many of these prisoners were housed at Marion based on discretionary determinations from administrative committees, which might have been arbitrary, improper, or dependent on confidential information which was never disclosed to the prisoner.173 The 1977 Supreme Court decision in Jones v. North Carolina Prisoners’ Labor Union, which equated any collective organizing, whether violent or not, with the potential for disruption, is illustrative of the Court’s use of the preventive rationale in the prison context.174 In Jones, the Court upheld officials’ prohibition of prisoners from soliciting other prisoners to join the union and of all union meetings. While the Secretary of the Department of Corrections stated that “[t]he purpose of the union may well be worthwhile projects,” and the district court held that there was “not one scintilla of evidence to suggest that the Union has been utilized to disrupt the operation of the penal institutions,” the Supreme Court nonetheless upheld the ban because this “historical finding . . . does not state that appellants’ fears as to future disruptions are groundless.”175 Defendants claimed that once a union was established, even if it had salutary purposes, “[w]ork stoppages and mutinies are easily foreseeable. Riots and chaos would almost

169 See Plaintiffs’ Motion for Extension of Settlement Agreement Based on Systemic Due Process Violations at 56–57, Ashker v. Brown, No. 4:09-cv-05796-CW (N.D. Cal. Nov. 20, 2017), available at https://ccrjustice.org/sites/default/files/attach/2017/11/Mot%20to%20Extend%20Settlement%20REDA CTED.pdf [https://perma.cc/7XP5-XFNN]. 170 Or, in the vernacular, to “parole, snitch, or die.” See Keramet A. Reiter, Parole, Snitch, or Die: California’s Supermax Prisons and Prisoners, 1997–2007, 14 PUNISHMENT & SOC’Y 530, 536 (2012). 171 New assignment guidelines were instituted when the BOP designated Marion as a Level 6, super- security prison in 1979. See supra note 93 and accompanying discussion. 172 WARD & BREED, supra note 74, at 35. 173 Id. 174 433 U.S. 119 (1977). 175 Id. at 127 & n.5.

191 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W inevitably result.”176 The Supreme Court found those fears rational and deferred to the expertise of the correctional officials.177 This same rationale permitted preventive administrative segregation for individuals who had committed no violent misconduct, but had the potential to do so in the future. Both banning unions and creating supermaxes were preventive measures designed to augment official control over prisoners. The replacement of typical punishment rationales with preventive and incapacitation strategies both drove the growth of supermax confinement and played a significant role in the rise of mass incarceration. As various scholars have noted, the rise of mass incarceration is associated with an increased reliance on penal incapacitation as a preventive measure.178 Starting in the 1970s, the criminal justice system’s focus shifted from punishing past crimes to the prevention of future misconduct by means of incarceration and the ongoing control exercised over supposedly dangerous offenders.179 The enactment of statutes such as “three strikes” laws, which authorize life sentences for repeat offenders; the criminalization of gang membership and recruitment; the continuation of detention for “sexual predators” beyond the service of the criminal sentences; and new sentencing guidelines which increased the sentence of offenders whose past histories allegedly make them most likely to commit future crimes, all reflect the shift from a focus on

176 Id. at 127. 177 Id. at 128–29. 178 Jonathon Simon, The Return of the Medical Model: Disease and the Meaning of Imprisonment from John Howard to Brown v. Plata, 48 HARV. C.R.-C.L. L. REV. 217, 239 (2013); see also BRUCE WESTERN, THE POLITICS AND ECONOMICS OF THE PRISON BOOM 37–38 (2005) (the prison boom was fueled by turning the penal system “to the twin tasks of incapacitation and deterrence”); FRANKLIN E. ZIMRING & GORDON HAWKINS, INCAPACITATION: PENAL CONFINEMENT AND THE RESTRAINT OF CRIME 8–12 (1995) (tracing shift from rehabilitation to incapacitation and its contribution to mass incarceration); Malcolm M. Feeley & Jonathan Simon, The New Penology: Notes on the Emerging Strategy of Corrections and Its Implications, 30 CRIMINOLOGY 449, 452 (1992) (“[T]he new penology is markedly less concerned with responsibility, fault, moral sensibility, diagnosis, or intervention and treatment of the individual offender. Rather, it is concerned with techniques to identify, classify, and manage groupings sorted by dangerousness.”); Miguel F.P. de Figueiredo, Throw Away the Key or Throw Away the Jail? The Effect of Punishment on Recidivism and Social Cost, 47 ARIZ. ST. L.J. 1017, 1027 (2015) (“In the United States, incapacitation became the predominant logic for the prison boom . . . .”); Jonathan Simon, Mass Incarceration: From Social Policy to Social Problem, in THE OXFORD HANDBOOK OF SENTENCING AND CORRECTIONS 23, 28 (Joan Petersilia & Kevin R. Reitz eds., 2012) (describing the abandonment of control in California in the 1970s and the increase in incarceration that followed, stating the change “abandoned the focus on rehabilitation in favor of punitive segregation intended to achieve deterrence and, more reliably, incapacitative effects”). 179 Robinson, supra note 25; Carol S. Steiker, The Limits of the Preventive State, 88 J. CRIM. L. & CRIMINOLOGY 771 (1998); Lucia Zedner, Seeking Security by Eroding Rights: The Side-Stepping of Due Process, in SECURITY AND HUMAN RIGHTS 257, 259–61 (Benjamin J. Goold & Liora Lazarus eds., 2007).

192 115:159 (2020) Mass Solitary and Mass Incarceration punishing criminal behavior to incapacitating dangerous individuals.180 Another manifestation of the preventive rationale is the continuation of the criminal justice system’s control over former prisoners and the debilitating consequences of conviction—ex-felons may be denied the right to vote, excluded from juries, and relegated to an often “racially segregated and subordinated existence.”181 Because the preventive paradigm often treats entire groups of people as dangerous based on their alleged characteristics, its prevalence has produced significant racial implications. For example, the dramatic rise in the use of the bail system, which jails mostly poor and African-American people accused of crime prior to their conviction primarily because of their presumed dangerousness, reflects the preventive model.182 Similarly, new policing tactics that fall heavily on Black and Latino communities, such as “stop and frisk” and “broken windows,” have as their main ideological underpinnings the targeting of so-called attributes of dangerousness as a preventive measure against crime.183 Inside prisons, states such as California and Texas have placed thousands of Hispanics into solitary confinement because of alleged association with gangs. The vagueness of the dangerousness criteria allowed ethnic identity to often become a proxy for gang association, which itself was a proxy for dangerousness.184

180 See DORIS LAYTON MACKENZIE, SENTENCING AND CORRECTIONS IN THE 21ST CENTURY: SETTING THE STAGE FOR THE FUTURE 9–11 (2001), https://www.ncjrs.gov/pdffiles1/nij/189106-2.pdf [https://perma.cc/GF7V-6F3M]; Jalila Jefferson-Bullock, How Much Punishment is Enough?: Embracing Uncertainty in Modern Sentencing Reform, 24 J.L. & POL’Y 345 (2016); Robinson, supra note 25, at 1429–31. 181 ALEXANDER, supra note 19, at 4; Sandra G. Mayson, Collateral Consequences and the Preventive State, 91 NOTRE DAME L. REV. 301 (2015). As Michelle Alexander has argued, these consequences are an important element of the New Jim Crow, which also has a preventive rationale. ALEXANDER, supra note 19, at 4. 182 Of the 2.3 million people incarcerated in 2019, more than 20% of them had not been convicted of a crime but were in preventive pretrial detention. Wendy Sawyer & Peter Wagner, Mass Incarceration, The Whole Pie 2020, PRISON POL’Y INITIATIVE (Mar. 24, 2020), https://www.prisonpolicy.org/reports/pie2020.html [https://perma.cc/3TYR-UGZR]. Over 555,000 people are locked up, mainly in local jails, who have not been convicted of a crime. Due to the high price of money bail, “people with low incomes are more likely to face the harms of pretrial detention.” Id.; see also Wendy Sawyer, How Race Impacts Who Is Detained Pretrial, PRISON POL’Y INITIATIVE (Oct. 9, 2019), https://www.prisonpolicy.org/blog/2019/10/09/pretrial_race/ [https://perma.cc/AWA8-JPUN]. 183 See, e.g., Floyd v. City of New York, 959 F. Supp. 2d 540, 562 (S.D.N.Y 2013) (noting the disproportionate stops of Blacks and Hispanics in holding that the use of stop and frisk without reasonable suspicion violated the Fourth Amendment and the Fourteenth Amendment). 184 See generally Bernard E. Harcourt, Risk as a Proxy for Race: The Dangers of Risk Assessment, 27 FED. SENT’G REP. 237 (2015) (arguing that risk assessment is an unacceptable tool that will exacerbate racial disparities in the criminal justice system as it has collapsed risk into prior criminal history, which has become a proxy for race).

193 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

The preventive rationale underlying both mass solitary and mass incarceration was also fueled by a sense of crisis and permanent emergency, which has been a prominent feature of post-World War II American society.185 As noted earlier, riots, disturbances, and other crises in prisons were traditionally met with temporary lockdowns or discrete periods of solitary confinement for prisoners who were disruptive. The marked change that began at Marion was to make the lockdown permanent, borne out of a sense that the crisis was not temporary. The supermax institutionalized the concept of permanent lockdown so that the struggle against gangs in prison required the indefinite, often permanent isolation of any prisoner deemed to be associated or affiliated with a gang. In that respect, the fight against gangs in prison can be analogized to the “war on terror” that the Bush Administration initiated. In response to a perceived existential crisis and a state of permanent emergency, the preventive paradigm played an important role in both mass efforts to detain and isolate individuals, with little due process, for their alleged (and often erroneous) association with dangerous groups.186 Courts have affirmed the constitutionality of permanent lockdowns in response to a perceived emergency.187 For example, in response to a threat from a designated group, the Fourth Circuit approved isolation for any alleged gang member, noting that in order to forestall a riot or other disturbance, prison officials may act without any showing that a particular individual is dangerous.188 According to the court, allowing prison officials to act only after a demonstration of individual dangerousness would deprive them of the “all-important option of prevention.”189 By the time of the Fourth Circuit’s decision, the prisoners had already been isolated for more than three years, simply due to their alleged affiliation, without any showing of individual dangerousness or of immediate threat of disorder.190 The perception of danger and threat had become permanent. Similarly, in Hewitt v. Helms, the Supreme Court allowed administrative segregation with only

185 See Jules Lobel, Emergency Power and the Decline of Liberalism, 98 YALE L.J. 1385, 1400–04 (1989) (describing the development of permanent emergency mentality, where conceptual crisis is permanent, and emergency authority, which was once seen as temporary, thereby becomes permanent). 186 See DAVID COLE & JULES LOBEL, LESS SAFE, LESS FREE: WHY AMERICA IS LOSING THE WAR ON TERROR 18–19 (2007). 187 See, e.g., Bruscino v. Carlson, 854 F.2d 162 (7th Cir. 1988). For a discussion on Bruscino, see supra notes 47–52 and accompanying text. 188 In re Long Term Admin. Segregation of Inmates Designated as Five Percenters, 174 F.3d 464, 470 (4th Cir. 1999). 189 Id. 190 Id. at 370–72.

194 115:159 (2020) Mass Solitary and Mass Incarceration minimal due process in response to a threat of riot,191 although, as Justice Stevens wrote in dissent, the emergency justification for such segregation only lasted a few days.192 Nonetheless, the inmate’s segregation continued even after the temporary emergency had ended.193 The distinction between emergency and normalcy had broken down; the supermax thus represented emergency normalized. Similarly, the rise of mass incarceration has been accompanied by a crisis mentality—illustrated by the terminology “war on drugs” or “war on crime”—initiated by President Lyndon Johnson over fifty years ago.194 A carceral instinct became a permanent feature of the crisis mentality characterized by these never-ending “wars.” As Professor David Garland notes, the criminal justice system has been in a “perpetual sense of crisis,” adding, however, that the term “crisis” seems “inappropriate for a situation that has now endured for several decades.”195 Moreover, that the high rates of incarceration, particularly of African Americans and Hispanics, have continued irrespective of actual crime rates suggests a mentality of perpetual crisis and emergency regardless of the existence of an imminent threat to society.196 Thus, the preventive model, premised on an ongoing sense of permanent emergency requiring the incapacitation of dangerous people who were usually nonwhites, underlay both the rise of mass incarceration and prolonged, mass solitary confinement.

IV. ENDING PROLONGED SOLITARY CONFINEMENT By the 1980s, courts had affirmed and legitimated prolonged solitary confinement under the theory that prison officials, charged with preventing violence and preserving order in prisons, had no alternative to the permanent or very extended mass lock-up of dangerous and violent prisoners.197 That rationale has three fundamental flaws, and understanding those flaws is essential to today’s movement for the reform and elimination of prolonged

191 459 U.S. 460, 472–76 (1983). 192 Id. at 489 n.16 (Stevens, J., dissenting). 193 Id. 194 Elizabeth Hinton, Why We Should Reconsider the War on Crime, TIME (Mar. 20, 2015), https://time.com/3746059/war-on-crime-history/ [https://perma.cc/FY7Y-XPJN] (noting that fifty years ago, President Johnson called for a “War on Crime”). 195 DAVID GARLAND, THE CULTURE OF CONTROL: CRIME AND SOCIAL ORDER IN CONTEMPORARY SOCIETY 19 (2001). 196 Andrew D. Leipold, Is Mass Incarceration Inevitable?, 56 AM. CRIM. L. REV. 1579, 1592–93 (2019) (“[T]here was a sustained drop in crime during the same period where there was a massive increase in the number of inmates—most of the 1990s and the first decade of the 2000s . . . .”). 197 See supra notes 47–55 (discussing the high-profile judicial decisions of Bruscino and Wilkinson).

195 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W solitary confinement. First, while judges such as Richard Posner understood that solitary confinement causes mental harm, later developments in the fields of psychology, neuroscience, and social science deepened the scientific consensus that prolonged solitary presents a profound risk of devastating mental and physical harm.198 Second, the widely held view in the 1980s and ’90s that supermax prisons were locking up the worst of the worst199 turns out to be false. Major prison systems such as California, Ohio, and New York have recognized that most of the prisoners preventively incarcerated in supermax or other prolonged solitary units did not require such draconian isolation and instead could have been managed in general population units.200 Finally, even for the relatively small number of truly dangerous, violent prisoners, who require separation from the general prison population, alternatives to the modern supermax have been proposed since the 1980s and continue to be developed today.201 Although viable alternatives have been proposed at critical junctures in the process of the “super- maximumization” of American prisons, these have largely been ignored by prison officials and legislatures.

A. Rejecting the Preventive Paradigm The preventive paradigm—whether used in the context of the “war on terror,” the “war on crime or drugs,” or the confinement of violent prisoners—generally leads to overclassification of so-called dangerousness

198 See generally SOLITARY CONFINEMENT, supra note 11 (containing chapters by social scientists, psychologists, and neuroscientists demonstrating harm to physical health caused by loneliness or lack of social interaction, mental harm and risk of social death, and potential harm to the brain caused by prolonged solitary confinement). 199 See DeMaio, supra note 143, at 210 (explaining that the benefit of removing the “most dangerous inmates” from the general prison population is the “general premise upon which supermax prisons are based”); Kurki & Morris, supra note 29, at 391 (“The new ‘dangerous’ prisoner is described as more violent, more disturbed, more disruptive, and, therefore, less likely to adjust to ordinary prison conditions . . . . Prison administrators often describe supermax inmates as ‘the worst of the worst[,]’ people who have nothing to lose and therefore do not hesitate from ‘taking a swing at a corrections officer or preying on another inmate.’” (citations omitted)); Maximilienne Bishop, Note, Supermax Prisons: Increasing Security or Permitting Persecution?, 47 ARIZ. L. REV. 461, 461 (2005) (“Super-maximum security (‘Supermax’) facilities are purported to house the most invidious and dangerous criminals in the nation’s prisons who pose such a threat to prison security that they can only be controlled by isolation.”). 200 Settlement Agreement at 4, Peoples v. Fischer, No. 1:11-cv-02694-SAS (S.D.N.Y. Dec. 16, 2015), available at https://www.clearinghouse.net/chDocs/public/PC-NY-0062-0011.pdf [https://perma.cc/9KPZ-BCF8]; Settlement Agreement at 10, Ashker v. Brown, No. C 09-05796 CW (N.D. Cal. Aug. 31, 2015), available at https://ccrjustice.org/sites/default/files/attach/2015/09/2015-09- 01-ashker-Settlement_Agreement.pdf [https://perma.cc/D6SS-DA44]. See infra note 217 and accompanying text on Ohio’s conversion of most of its supermax prison to a maximum-security prison after the district court decision in Austin v. Wilkinson. 201 See infra Section I.B.

196 115:159 (2020) Mass Solitary and Mass Incarceration without clear standards or due process to restrict the state in whom it detains.202 That is exactly what happened in the prolonged solitary confinement context. An alternative urged at the time of the Marion lockdown was to discard the preventive model and return to a system that addresses crisis and misconduct through temporary lockdowns. This system would discipline, for a determinate period, only those prisoners who had engaged in serious misconduct, as opposed to preventively locking down the entire prison permanently.203 California and Ohio are examples of the enormous overclassification of supposedly dangerous prisoners. California, as already mentioned, placed thousands of alleged gang members or associates in supermax SHU prisons with no way out. In the context of a lawsuit brought by a class of prisoners held in Pelican Bay SHU, James Austin, a corrections classification expert, submitted a report finding that California’s use of a status-based system relying on gang affiliation for placement and retention in the SHU results in a system whereby individuals who actually present no major management problem are retained in SHU for “excessive periods of time.”204 Austin’s conclusions were shared by Emmitt Sparkman, a longtime correctional official who had overseen the reforms of Mississippi’s solitary confinement unit.205 Austin also discovered that the California Department of Corrections and Rehabilitation (CDCR) never examined whether or not SHU placement of gang-affiliated inmates reduces violence throughout its prison system.206

202 See COLE & LOBEL, supra note 186, for the general problems attendant to the preventive paradigm. 203 See, e.g., Oversight Hearing, supra note 16, at 107 (congressional testimony of Jan Susler) (citing cases and wardens challenging long-term lockdowns); WARD & BREED, supra note 74, at 47 (testimony of David Ward) (arguing that lockdown should not be permanent). 204 Expert Report of Dr. James Austin at 2, 7–8, Ashker v. Brown, No. 4:09-cv-05796-CW (N.D. Cal. Mar. 13, 2015), available at https://ccrjustice.org/sites/default/files/attach/2015/07/Redacted_ Austin%20Expert%20Report.pdf [https://perma.cc/RSW9-QSBH]. Austin studied the records of the named plaintiffs and found that they had “an exceptionally low rate of disciplinary infractions for a ten- year period for a high security population.” Id. at 15. The vast majority of those violations were minor, such as unauthorized talking. For Austin, “[a] system that places such inmates in SHU for over a decade defies all logic.” Id. He concluded that “[t]he inmate classification and disciplinary conduct data all suggest that these inmates, in general, do not require SHU placement.” Id. at 19. Indeed, Austin’s review of forty-one plaintiff class members found that over 70% were assessed by the CDCR as “low risk,” a designation seemingly in contradiction to their continued placement in the SHU. Id. at 20. 205 Expert Report of Emmit L. Sparkman at 3, 17–19, Ashker v. Brown, No. 4:09-cv-05796-CW (N.D. Cal. Mar. 5, 2015), available at https://www.clearinghouse.net/chDocs/public/PC-CA-0054- 0023.pdf [https://perma.cc/S72L-V5LP] (“There is no penological justification for long term special housing of offenders for investigation, protective custody, non-violent rule violations, length and/or type of prison sentence. . . . [T]he CDCR places and retains more offenders [than other prison systems] without safety and security justification for longer periods of time with harsher conditions.”). 206 See Expert Report of Dr. James Austin, supra note 204, at 11.

197 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

When Austin conducted that review himself, he found that the increased use of SHU had not produced lower assault rates in CDCR prisons, but rather, the rate of assault increased.207 After more than twenty-five years, CDCR finally admitted that its SHU policy of preventively keeping prisoners associated with gangs in prolonged solitary confinement “was a mistake”208 and settled the Ashker v. Brown lawsuit, agreeing to no longer place prisoners in indeterminate solitary confinement based on gang status but only for proven serious misconduct after a due process hearing resulting in a determinate SHU sentence.209 While the determinate SHU sentences prisoners can receive are still extensive, 210 and CDCR continues to send many individuals to the SHU based on unreliable or fabricated confidential information,211 California is placing far fewer prisoners in solitary confinement and has converted a wing of the infamous Pelican Bay SHU into a minimum security general population unit.212 Similarly, Ohio’s practices illustrate the use of the supermax to house hundreds of prisoners who do not require such high security. Ohio built its supermax in Youngstown in response to the 1993 Lucasville Prison Riot and started placing prisoners there in 1998.213 The Lucasville riot was undoubtedly brought on by overcrowding, faulty prison management, and a failure to respond to legitimate prisoner grievances and peaceful protests,

207 Id. at 20. 208 Oprah Winfrey, Reforming Solitary Confinement at an Infamous California Prison, 60 MINUTES (July 22, 2018), https://www.cbsnews.com/news/60-minutes-reforming-solitary-confinement-at-an- infamous-california-prison/ [https://perma.cc/VA36-5UT2] [hereinafter 60 MINUTES] (Scott Kernan, Secretary of CDCR, states that the policy of sending gang members to SHU indefinitely “was a mistake . . . . It didn’t work because of the impact on the offenders.” (emphasis omitted)); see also Expert Report of Dr. James Austin, supra note 204, at 13 & n.18 (noting that the CDCR has explicitly acknowledged that its policy “overclassified” prisoners for SHU placement). 209 Settlement Agreement, Ashker v. Brown, supra note 200, at 7–8. 210 A gang member who commits murder in prison related to gang activities can still theoretically receive a five-year SHU sentence with two additional years in a step-down program, although in practice the sentences have been considerably shorter. See id. at 12. 211 Ashker v. Newsom, No. 09-cv-05796-CW, 2019 U.S. Dist. LEXIS 13382, at *51 (N.D. Cal. Jan. 25, 2019) (finding that California’s use of confidential information to place people in the SHU systemically violated the Due Process Clause of the Fourteenth Amendment). 212 Between December 2012 and August 2016, California’s entire solitary confinement population dropped from 9,870 to 3,471. California Solitary Confinement Statistics: Year One After Landmark Settlement, CTR. FOR CONST. RTS. (Oct. 18, 2016), https://ccrjustice.org/california-solitary-confinement- statistics-year-one-after-landmark-settlement [https://perma.cc/Q3LU-F4YH]; see also 60 MINUTES, supra note 208 (Scott Kernan discussing converting a wing of the Pelican Bay SHU into a minimum- security unit). 213 Austin v. Wilkinson, 189 F. Supp. 2d 719, 722–23 (N.D. Ohio 2002); see also LYND, supra note 56, at 11.

198 115:159 (2020) Mass Solitary and Mass Incarceration including a class action lawsuit challenging the overcrowding and double- celling which was eventually dismissed by the United States Supreme Court.214 In 2001, a prisoner class action lawsuit challenged conditions at Ohio’s Supermax Prison (OSP) and the State’s procedures for placement and retention there. Judge James Gwin found that [o]pened in 1998, the OSP is an ill-conceived legislative remedy to a problem that did not exist. Reacting to the horrendous April 1993 riot at the Southern Ohio Correctional Facility at Lucasville, the General Assembly poured huge amounts of state funds into OSP, Ohio’s first supermax prison. The fault of this plan lies in the fact that the Lucasville riot was caused by overcrowding in the maximum security area, not by any lack of space in the high maximum security. Despite a need for maximum security cells, the Ohio General Assembly built OSP to provide high-maximum security cells, cells for which there was little need.215 The district court in Austin found that the procedures for placement and retention at OSP violated due process and that numerous prisoners had been placed and retained there inappropriately.216 It ordered new hearings, which resulted in a dramatic reduction in the number of prisoners housed in the supermax and led Ohio to convert most of the supermax into a maximum (not super-maximum) security facility.217 Ohio and California’s supermax experiences illustrate that mass solitary, as with mass incarceration, incapacitated and isolated numerous

214 Rhodes v. Chapman, 452 U.S. 337 (1981); LYND, supra note 56, at 12–30 (describing the conditions and the peaceful protests by prisoners attempting to change the situation). In Rhodes, prisoners contended that the close confinement of double-celling for long periods creates a dangerous potential for frustration, tension, and violence, which could lead to rioting, a concern rejected by the Court’s majority as a basis for an Eighth Amendment violation. Id. at 349 n.14. More than ten years later, the prisoners did riot, and a report by a commission of prison officials led by Gary Mohr reported that “double celling of the inmate population was voiced by a vast majority of both staff and inmates as a cause of the disturbance.” LYND, supra note 56, at 23. 215 Order, Austin v. Wilkinson, No. 4:01-CV-71 (N.D. Ohio Sept. 30, 2005), ECF No. 624; see also Austin, 189 F. Supp. at 723 (noting that evidence suggested that Ohio did not need a supermax prison). 216 Austin, 189 F. Supp. 2d at 749. 217 Austin v. Wilkinson, 502 F. Supp. 2d 660, 674–75 (N.D. Ohio 2006); Brief for Respondents at 17–18 & n.5, Wilkinson v. Austin, No. 04-495, 2005 WL 556835 (Mar. 4, 2005). Ohio’s statistics demonstrate that by the end of 2004 when the district court’s procedures had been fully implemented, the population of the supermax declined from over 400 prisoners at the outset of the lawsuit to 55 prisoners remaining in Level V, or super-maximum classification. E-mail from Alice Lynd to Jules Lobel (Aug. 9, 2019, 7:00 PM) (on file with journal).

199 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W people who did not warrant such isolation.218 As with mass incarceration, there is a tremendous reform movement even amongst prison officials who recognize that mass solitary has resulted in the isolation, incapacitation, incarceration, and essentially the discarding of thousands of people whose misconduct does not warrant such treatment.219 Society and the courts are rediscovering the lesson, apparent in the 1800s, that solitary confinement wreaks profound damage to a person’s psychological state,220 and modern neuroscience and social science have now recognized that isolation leads to tremendous damage to the brain and body as well.221 Moreover, as a recent National Academy of the Sciences landmark study on mass incarceration concluded, the use of supermax and other extreme forms of isolation in America’s prisons has “done little or nothing to reduce system-wide prison

218 See, e.g., ALEXANDER, supra note 19 (arguing that mass incarceration has led to the incarceration of hundreds of thousands of minorities who did not warrant incarceration); Alexander, supra note 33 (same). 219 See, e.g., TIME-IN-CELL, supra note 13, at iii; Gottschalk, supra note 14, at 261–63 (highlighting the support for reform of solitary confinement by corrections officials); Timothy Williams, Prison Officials Join Movement to Curb Solitary Confinement, N.Y. TIMES (Sept. 2, 2015), https://www.nytimes.com/2015/09/03/us/prison-directors-group-calls-for-limiting-solitary- confinement.html [https://perma.cc/WP5J-M5YE]; see generally Amy Fettig & David C. Fathi, “Loneliness Is a Destroyer of Humanity”, in SOLITARY CONFINEMENT, supra note 11, at 343 (describing the current reform movement against prolonged solitary confinement); Alexander A. Reinert, Solitary Troubles, 93 NOTRE DAME L. REV. 927 (2018) (discussing current reform movement). 220 See, e.g., Porter v. Clarke, 923 F.3d 348, 356–57 (4th Cir. 2019) (noting the psychiatric risks of solitary confinement); Williams v. Sec’y Pa. Dep’t of Corr., 848 F.3d 549, 566–69 (3d Cir. 2017) (accepting the “scientific consensus” that solitary confinement “‘is psychologically painful, can be traumatic and harmful, and puts many of those who have been subjected to it at risk of long-term . . . damage’” (quoting Haney & Lynch, supra note 27, at 500)), cert. denied sub nom. Walker v. Farnan, 138 S. Ct. 357 (2017); Craig Haney, Mental Health Issues in Long-Term Solitary and “Supermax” Confinement, 49 CRIME & DELINQ. 124, 132 (2003) (summarizing the numerous psychological studies that demonstrate the damage to the individual wrought by solitary confinement, stating that “there is not a single published study of solitary or supermax-like confinement in which nonvoluntary confinement lasted for longer than 10 days, where participants were unable to terminate their isolation at will, that failed to result in negative psychological effects”). 221 See, e.g., Louise Hawkley, Social Isolation, Loneliness, and Health, in SOLITARY CONFINEMENT, supra note 11, at 185 (summarizing social science evidence that isolation harms the physical health of the individual); Jules Lobel & Huda Akil, Law & Neuroscience: The Case of Solitary Confinement, 147 DAEDALUS 61 (2018) (summarizing neuroscientific evidence that solitary confinement damages the brain).

200 115:159 (2020) Mass Solitary and Mass Incarceration disorder or disciplinary infractions.”222 As a result, courts have begun to impose significant restrictions on the use of solitary confinement.223

B. Alternatives to Solitary Confinement for Violent Prisoners As with mass incarceration, recent judicial decisions and reform efforts still have not adequately addressed the problem of the very violent individual who most prison officials would say requires prolonged solitary confinement to manage. What Danielle Sered and others have pointed out in the mass incarceration context is relevant to the solitary confinement reform movement: “The current reform narrative, though compelling, has been based on a fallacy; that the United States can achieve large-scale transformative change . . . by changing responses to nonviolent offenses.”224 Both mass incarceration and mass solitary have arisen based on a false and racist narrative of the violent individual: At the heart of that narrative is the story of an imagined monstrous other—a monster who is not quite human like the rest of us, who is capable of extraordinary harm and incapable of empathy, who inflicts great pain but does not feel it as we do, a monster we and our children have to be protected from at any price.225 The image of this alleged monster perpetuated the rise of mass solitary confinement as much as, if not more, than mass incarceration itself.226 Yet,

222 COMM. ON CAUSES & CONSEQUENCES OF HIGH RATES OF INCARCERATION, THE GROWTH OF INCARCERATION IN THE UNITED STATES: EXPLORING CAUSES AND CONSEQUENCES 186 (Jeremy Travis, Bruce Western & Steve Redburn eds., 2014); see also Atul Gawande, Hellhole, NEW YORKER (Mar. 23, 2009), https://www.newyorker.com/magazine/2009/03/30/hellhole [https://perma.cc/2GHK-TTVX] (citing 2003 study finding that after supermax prisons were opened in Arizona, Illinois, and Minnesota, “levels of inmate-on-inmate violence were unchanged”). 223 See Porter, 923 F.3d at 368; Williams, 848 F.3d at 572–74; Settlement Agreement, Ashker v. Brown, supra note 200, at 4–16; see generally Fettig & Fathi, supra note 219 (describing the current reform movement against prolonged solitary confinement). 224 SERED, supra note 33, at 5–6; see also Alexander, supra note 33 (“Despite the abysmal failure of ‘get tough’ strategies to break cycles of violence in cities like Chicago, reformers of our criminal justice system in recent years have largely avoided the subject of violence, instead focusing their energy and resources on overhauling our nation’s drug laws and reducing penalties for nonviolent offenses.”). 225 SERED, supra note 33, at 11. 226 Justin Peters, How a 1983 Murder Created America’s Terrible Supermax-Prison Culture, SLATE (Oct. 23, 2013, 3:56 PM), https://slate.com/news-and-politics/2013/10/marion-prison-lockdown-thomas- silverstein-how-a-1983-murder-created-america-s-terrible-supermax-prison-culture.html [https://perma.cc/BN3Q-8XD7]. In 1998, former Federal Bureau of Prisons Director Norman Carlson justified his decision to lock down Marion as a harsh but necessary and effective measure, telling the San Francisco Chronicle that “[t]here is no way to control a very small subset of the inmate population who show absolutely no concern for human life . . . . [Silverstein] . . . had multiple life sentences. Another life sentence is no deterrent.” Id. (emphasis added); see generally JAMES KILGORE, UNDERSTANDING MASS

201 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W prolonged solitary confinement’s extremely deleterious effects on an individual and its affront to human dignity constitutes torture as well as cruel and degrading treatment prohibited by the Eighth Amendment and international law.227 As torture, it should be prohibited generally, and not simply against those who are nonviolent or not violent enough to supposedly warrant such treatment. We will never end prolonged solitary confinement, nor recognize its true nature as torture of the body and soul, unless we develop alternative, humane ways of treating those few whose persistent violence does require some restrictions and separation from the general prison population. The 1983 Marion permanent lockdown prompted suggestions of alternatives to draconian isolation that were ignored and long forgotten. In 1985, the House of Representatives Subcommittee on Courts, Civil Liberties and the Administration of Justice, concerned about the conditions of solitary confinement at Marion, held oversight hearings into the lockdown at Marion.228 The subcommittee commissioned two consultants with substantial expertise and experience in the field of corrections, David Ward and Allen Breed, to investigate the situation at Marion and make recommendations about what to do moving forward. Ward and Breed’s “most important recommendation” was that the lockdown not be accepted as a permanent institution.229 As a long-term alternative for managing the highest security prisoners in the system, they urged the “[c]onstruction of a ‘new generation’ level 6 prison along the lines of the Minnesota Correctional Facility at Oak Park Heights.”230 In their opinion, such a prison would combine the high-tech surveillance features of a modern supermax with small units that would have “40–50 inmates, all in individual cells, contain[] dining and laundry areas, counselling offices,

INCARCERATION: A PEOPLE’S GUIDE TO THE KEY CIVIL RIGHTS STRUGGLE OF OUR TIME 35–36 (2015); Shawn E. Fields, Weaponized Racial Fear, 93 TUL. L. REV. 931, 944–47 (2019) (tracing the effects of criminal justice policies on the proliferation of the “black criminal underclass trope”); Joshua Kleinfeld, Two Cultures of Punishment, 68 STAN. L. REV. 933, 1008–15 (2016) (considering a philosophical approach relating perceptions of evil characteristics to ideologies on punishment); Jonathan Simon, Is Mass Incarceration History?, 95 TEX. L. REV. 1077, 1089–94 (2017) (providing a historiographical perspective addressing strategies targeting perceived racial threats in the rise of mass incarceration). 227 See UNITED NATIONS OFFICE ON DRUGS & CRIME, THE UNITED NATIONS STANDARD MINIMUM RULES FOR THE TREATMENT OF PRISONERS (THE NELSON MANDELA RULES) 13–14 (2015) https://www.unodc.org/documents/justice-and-prison-reform/Nelson_Mandela_Rules-E-ebook.pdf [https://perma.cc/S3GQ-54DE] (prohibiting indefinite and prolonged solitary confinement and defining prolonged solitary confinement as being more than fifteen days in solitary). 228 See Oversight Hearing, supra note 16. 229 WARD & BREED, supra note 74, at 47. 230 Id. at 27.

202 115:159 (2020) Mass Solitary and Mass Incarceration indoor game rooms, a wire enclosed outdoor recreation yard and a work area.”231 The relatively small, self-contained units would “allow congregate activities on a unit basis.”232 In sum, while a prison in the Oak Park Heights model could be considered a supermax, it would not have the distinguishing features of a modern supermax: extreme social isolation, lack of environmental stimulation, and absence of physical contact. A study conducted by Professor Roy King of the prison conditions at Oak Park Heights confirmed the consultants’ observations. Prisoners were permitted to eat in small groups at tables if they wanted, received up to sixteen hours per month of contact visits, engaged in small group recreation and work opportunities, and had access to fifteen- or thirty-minute phone calls, which about 30% of the prisoners used daily.233 Had the Oak Park Heights model been followed by the federal government and other states, it might have changed the course of the supermax boom that followed.234 Oak Park Heights, at least initially, was very secure and safe by modern prison standards. Remarkably, after ten years of operation, the prison had experienced no escape attempts or serious acts of violence, despite housing very dangerous prisoners serving life without parole, including about twenty-five to thirty prisoners who had been transferred from Marion.235 One former warden of Oak Park Heights attributes its success to tight security combined with a positive attitude towards prisoners, which allowed high- risk prisoners out of their cells for most of the day, provided significant programming, and attempted to create a positive environment, unlike the traditional supermax.236 Former Warden Bruton wrote in 2004 that after

231 Id. at 28. 232 Id. 233 Roy D. King, Maximum-Security Custody in Britain and the USA: A Study of Gartree and Oak Park Heights, 31 BRITISH J. CRIMINOLOGY 126, 146–47 (1991); cf. JAMES H. BRUTON, THE BIG HOUSE: LIFE INSIDE A SUPERMAX SECURITY PRISON 41 (2004) (stating that more than 50% of the prisoners housed at Oak Park Heights had killed someone, 95% are violent offenders, and a “significant number are serving life without parole”). 234 See King, supra note 5, at 172–73 (contrasting Oak Park Heights with supermaxes and noting that substantial programming at the former and the absence of such programming in the supermaxes was neither “a necessary [n]or desirable feature”). 235 BRUTON, supra note 233, at 150. Like Marion, Oak Park Heights was built in response to violence and disturbances that wracked the state’s maximum-security prison (Minnesota Correctional Facility at Stillwater). Id. at 27–28. 236 Id. at 31–41; see also King, supra note 233, at 147–48 (stating that it is well known that the success of Oak Park Heights is “in large part due to the management philosophy of its exceptional Warden” who embraces a philosophy whereby staff “should treat inmates as we would want our sons, brothers or fathers treated”); see also CLEMENS BARTOLLAS, BECOMING A MODEL WARDEN: STRIVING FOR EXCELLENCE (2004) (describing the life and philosophy of Frank Wood, the first warden at Oak Park Heights).

203 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W twenty years of operation there had not been a homicide or escape, nor were drugs or homemade weapons rife at Oak Park Heights.237 Of course, the prison was never problem free, and conditions may have become far worse since its first decade of existence. Today, it has a draconian disciplinary segregation unit holding fifty-two prisoners in what can for some be very prolonged solitary. Offenders who do not follow the rules are sometimes threatened with transfer to isolation in the federal system, and physical contact with visitors has been severely limited.238 Moreover, there has recently been a spike in violence at the prison, with instances of assault on staff and prisoners rising.239 Yet, Oak Park Heights’s practices, at least in its beginnings, of confining dangerous prisoners in high security, but not isolating, conditions serves as a potential alternative to the prolonged solitary confinement of modern supermax confinement. The Federal Bureau of Prisons rejected Ward and Breed’s suggestion. Initially, the BOP decided it did not need a new supermax prison, noting that the number of prisoners needing such high security was not increasing. The BOP concluded it would be a more efficient use of limited resources to construct “additional medium security facilities” to “reduce the overall level of crowding” than to put resources into a supermax for a “small number of inmates.”240 Nonetheless, within a few years, the BOP reversed course, and decided that it did, after all, need a supermax.241 But instead of revisiting the model that Ward and Breed suggested242—a high security, tightly controlled

237 BRUTON, supra note 233, at 38–39. 238 Id. at 52, 86–87, 103. 239 Bob Shaw, Violence Spikes at Oak Park Heights Prison, and the Search for Causes Is Urgent, TWIN CITIES PIONEER PRESS (Apr. 18, 2018, 7:26 PM), https://www.twincities.com/2018/04/18/as- violence-spikes-at-oak-park-heights-prison-the-search-for-solutions-grows-urgent/ [https://perma.cc/5RFW-JGLG]. 240 Oversight Hearing, supra note 16, at 150–51 (Letter from BOP Director Norman Carlson to Congressman Robert W. Kastenmeier (Jan. 7, 1985)). 241 See FROM ALCATRAZ TO MARION TO FLORENCE, supra note 60, at 5. As Professor Reiter has shown in other contexts, the decision to build ADX was not made the subject of public or even legislative discussion, but was shielded from public view. REITER, 23/7, supra note 14. For example, a 1990 FOIA request for information on the pending plans to construct the ADX was denied on the grounds that no such plans existed. FROM ALCATRAZ TO MARION TO FLORENCE, supra note 60, at 13 n.7 (citing Letter from Wallace H. Cheney, Gen. Counsel for the Fed. Bureau of Prisons, to Jan Susler, Att’y for the People’s Law Office (Dec. 31, 1990)). 242 Indeed, the then-Director of the BOP testified before Congress in 1989 that the BOP was planning to build a new high-security unit that would function similarly to Oak Park Heights: [W]e are designing right now and hope to build with funds in 1990 a state-of-the-art facility for [L]evel 6 prisoners in the Bureau of Prisons that would replace Marion, that would allow us to permit prisoners more freedom during the day within a unit similar to a facility that was operated

204 115:159 (2020) Mass Solitary and Mass Incarceration prison, which nonetheless allowed for group recreation in small units, congregate meals, work opportunities, phone calls, and contact visits with family and friends—the BOP constructed a new, more draconian version of the Marion lockdown model, the ADX Supermax at Florence.243 As public opinion, judicial decisions, and prison administrators have reevaluated the use of prolonged solitary confinement in American prisons, there has been increased interest in a model of segregating dangerous prisoners without isolation, such as the one that Oak Park Heights initially presented. Former Colorado Department of Corrections Director Rick Raemisch has led the way in imagining and implementing alternative approaches, even for the very violent prisoner, as has Director Leann Bertsch in North Dakota.244 Raemisch eliminated mass prolonged solitary confinement in Colorado and set up alternative units relying on small-group, congregate activities to house dangerous prisoners.245 Some European nations have also developed alternatives that segregate high-risk prisoners without the harsh social isolation found in American supermax prisons. In Scotland, England, and Wales, for example, dangerous prisoners are confined away from the general population, but in small groups rather than total isolation.246 There they are provided direct-contact family and legal visits and telephone calls, as well as “access to education, gym facilities, payment for work, association with other prisoners, and in-cell activities.”247 Perhaps even more striking is the prison at Grendon in England, which houses some of the most “damaged, disturbed and dangerous” prisoners in the English prison system.248 Despite its difficult population, Grendon provides small-group therapy and daily community meetings and has produced, in the words of its governor, “extraordinary outcomes.”249

a few years ago in Oak Park Heights in Minnesota that provides security and yet at the same time allows a greater access to programs. Oversight Hearings on Corrections: Overcrowding and Alternatives to Incarceration Before the Subcomm. on Courts, Intellectual Property & the Admin. of Justice of the H. Comm. on the Judiciary, 101st Cong. 34 (1989). 243 See FROM ALCATRAZ TO MARION TO FLORENCE, supra note 60, at 11–13. 244 See Bertsch, supra note 42, at 325; Raemisch, supra note 42, at 311–13. 245 The author visited these units in August 2018. 246 See Expert Report of Dr. Andrew Coyle ¶¶ 37, 44, Exhibit 2, Ashker v. Brown, No. 4:09-cv- 05796-CW (N.D. Cal. Mar. 12, 2015), available at http://ccrjustice.org/sites/default/files/attach/2015/07/Coyle%20Expert%20Report.pdf [http://perma.cc/P7WJ-DRVG]. 247 Id. ¶ 37. 248 Jamie Bennett, Resisting Supermax: Rediscovering a Humane Approach to the Management of High-Risk Prisoners, in SOLITARY CONFINEMENT, supra note 11, at 279, 287. 249 Id. at 289.

205 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

Integral to the separation-without-isolation model is a humane approach to prison management. As numerous prison experts pointed out at the time and continue to urge, the violence and disturbances at Marion and other prisons in the 1970s and ’80s were undoubtedly at least in part brought on by inhumane, hostile prison management which did not recognize or adequately respond to legitimate grievances.250 The ACLU and other prison experts testified before the Congressional Committee and before the courts that the practices of the prison officials both before and during the lockdown should be evaluated to determine whether “the errors or weaknesses in the prison administration . . . created a climate for the occurrence of violence or which exacerbated the violent confrontation.”251 As numerous correctional officials have recognized, treating prisoners humanely and responding to legitimate grievances are key mechanisms in tamping down violence in prisons.252 The court in Bruscino nevertheless categorically rejected any notion that changing prison administration practices would at least help reduce the violence at Marion and other prisons.253 And neither the BOP nor Congress ever undertook any investigation into whether prison management practices either led to or failed to prevent the violence that occurred at Marion in 1983. Once we remove the thousands of nonviolent or mentally ill prisoners from supermaxes and other prolonged solitary units, it will be possible to develop more positive, intensive programs incorporating social interaction to house those who truly do require some separation from the general prison

250 See, e.g., THOMPSON, supra note 56, at 1–50 (describing the peaceful attempts of the Attica prisoners to get redress for their grievances); N.Y. STATE SPECIAL COMM’N ON ATTICA, ATTICA: THE OFFICIAL REPORT OF THE NEW YORK STATE SPECIAL COMMISSION ON ATTICA 106–08 (1972) (same). 251 Oversight Hearing, supra note 16, at 60–61 (statement of ACLU National Prison Project) (calling for an investigation of the prison administration practices that may have helped cause the violence). As the ACLU statement noted, there was a significantly higher rate of violence and homicides in the federal prison system than in comparable state systems, even though the federal system has a higher budget, newer facilities, and no more violent prisoners or gang violence than in comparable state systems. Id. at 63 n.11; see also id. at 67–68 (declaration of Vincent Nathan) (statement from well-respected expert in corrections who submitted report to the committee stating that “the level of violence at a facility can be significantly decreased when the complaints of prisons are dealt with fairly”); id. at 109, 125, 130 (report from Marion Prisoners’ Rights Project citing testimony of Craig Haney, Bernard Rubin, and Dr. Arnold Abrams that management of control unit, which dehumanizes prisoners, breeds anger and violence). 252 See Expert Report of Dr. Andrew Coyle, supra note 246, at 16, 21 (noting that the most effective means of prison management require positive security in which staff relate to prisoners). 253 Bruscino v. Carlson, 854 F.2d 162, 166 (7th Cir. 1988) (“[P]laintiffs argue that it is the conditions at Marion before the lockdown that brutalized the inmates and caused them to become so violent and destructive. This is rank conjecture and implausible to boot. Few inmates are assigned to Marion who do not have a substantial history of violence in prison; it is not likely that these wolves would have turned into sheep if Marion had been a gentler place.”). The history of Oak Park Heights, holding dangerous prisoners who also had a history of violence, would contradict the Bruscino court.

206 115:159 (2020) Mass Solitary and Mass Incarceration population. And authorities should rediscover the Oak Park Heights model as constructed by “model warden” Frank Wood in the 1980s as an alternative to the modern supermax, even for very dangerous prisoners.254 As congressional consultant David Ward put it, “The challenge for the Bureau of Prisons, as we tried to emphasize in our report, is to try to do something positive even with those problematic individuals under these very special circumstances.”255 At the heart of ending both mass incarceration and prolonged solitary confinement lies the quest to treating even very violent people in a “positive,” humane way. This will preserve their human dignity and reflect the rehabilitative goals of punishment which are inherent in the value that even those who have committed terrible acts are capable of redemption and change.256 The conditions at modern supermaxes reflect a mission not of protecting against violence, but of exercising absolute control over prisoners, a goal that is facilitated by debilitating prisoners’ psychological states. The first step towards an alternative to the mass use of solitary is to separate punitive control from legitimate security functions. From a security perspective, it is difficult to perceive any benefit to providing small, enclosed recreation areas without equipment, windowless cells, and no phone calls with family and friends, as was the case at Pelican Bay SHU.257 The absolute deprivation of reading materials, affirmed by the Supreme Court in Beard v. Banks as “logical,”258 serves no serious security goal, nor does the widespread censorship of books and materials,259 nor the prohibition of media interviews. In contrast, “positive” security, which involves the kind of engaged staff–prisoner interactions banished from the supermax, can be combined with small-group social interaction to provide human contact, even for very dangerous prisoners.260 Perhaps most fundamentally, a recognition that the rise of the supermax was undergirded by a perceived need to exercise total control over prisoners deemed disruptive requires that reforms accord even allegedly dangerous

254 See BARTOLLAS, supra note 236. 255 WARD & BREED, supra note 74, at 48. 256 See, e.g., BRYAN STEVENSON, JUST MERCY (2014). 257 See, e.g., Madrid v. Gomez, 899 F. Supp. 1146, 1266 (N.D. Cal. 1995) (“[S]ubjecting individuals to conditions that are ‘very likely’ to render them psychotic or otherwise inflict a serious mental illness or seriously exacerbate an existing mental illness [cannot] be squared with evolving standards of humanity or decency, especially when certain aspects of those conditions appear to bear little relation to security concerns.”). 258 548 U.S. 521, 531–32 (2006). 259 See RACHAEL KAMEL & BONNIE KERNESS, THE PRISON INSIDE THE PRISON: CONTROL UNITS, SUPERMAX PRISONS, AND DEVICES OF TORTURE 2 (2003), https://www.afsc.org/sites/default/ files/documents/PrisonInsideThePrison.pdf [https://perma.cc/DA7J-SR32]. 260 See Expert Report of Dr. Andrew Coyle, supra note 246, at 17.

207 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W prisoners with some ability to dialogue with officials over the conditions under which they live and challenge unjust policies. In the California litigation that ended indeterminate solitary confinement in that state, the most difficult aspect of the settlement was not resolving disputes over the substantive policies that the State would henceforth implement, but rather plaintiffs’ insistence that CDCR officials should meet with their representatives on a quarterly basis to discuss the implementation of the settlement decree. Those officials were willing to meet with plaintiffs’ lawyers but not with the plaintiffs themselves, perhaps believing that to do so would accord prisoner representatives legitimacy and undermine officials’ total control. Eventually it required mediation by the federal judge overseeing the process to get CDCR to accept a compromise of semiannual meetings between CDCR officials and prisoner representatives.261 When those meetings occurred after the Agreement was implemented, CDCR officials often refused to even engage in a dialogue with the prisoner representatives.262 Yet the challenge of according individuals confined in high-security units some collective control over their circumstances and the conditions of their confinement is critical to restoring human dignity and hope to those who suffered through years of solitary confinement.

CONCLUSION The current reform movement to end prolonged solitary confinement has two major tasks. The first is to demonstrate the risk of harm the draconian practice imposes on those prisoners subjected to it. In the last decade, our understanding of the harm wrought by solitary confinement has been deepened by the work of social scientists who have shown that loneliness and social isolation are as great a risk factor for a number of serious physical conditions, such as hypertension, heart attacks, and strokes, as smoking or obesity.263 The work of neuroscientists has also established that solitary confinement harms the human brain.

261 See Settlement Agreement, Ashker v. Brown, supra note 200, ¶ 49 (“Defendants shall meet with Plaintiffs’ counsel and the four inmate representatives semiannually to discuss progress with implementation of this Agreement.”). 262 Letter from Jules Lobel to Judge Vadas (Dec. 2, 2016) (on file with journal). 263 See Expert Report of Louise C. Hawkley at 4–8, Ashker v. Brown, No. 4:09-cv-05796-CW (Mar. 12, 2015) (discussing health effects of social isolation); Hawkley, supra note 221, at 185–98 (summarizing social science evidence that isolation harms the physical health of the individual); Julianne Holt-Lunstad, Timothy B. Smith & J. Bradley Layton, Social Relationships and Mortality Risk: A Meta- Analytic Review, 7 PLOS MED., no. 7, 2010, at 14–15, available at https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1000316 [https://perma.cc/SKX4-6KXW] (finding correlation between social interaction and mortality that is comparable to other health risks).

208 115:159 (2020) Mass Solitary and Mass Incarceration

The second task is to overcome the mythology of the violent predator, for whom prison officials have no alternative but to confine in draconian isolation from other inmates, staff, and even families and friends. This Essay has focused on that task. It has demonstrated that mass solitary in this country developed not in response to that violent predator, but rather as a means for officials to achieve control of political activists and “troublemakers” amongst prisoners. Moreover, most prisoners caught up in the solitary dragnet could be safely managed in well-run general population units instead of warehoused in modern supermaxes. Finally, there is an alternative to the modern supermax for those few who cannot be safely managed in general population: separation without isolation.

209 N O R T H W E S T E R N U N I V E R S I T Y L A W R E V I E W

210